Clinicia is comprehensive cloud based SaaS Practice management Software which enables Doctors to update and manage their Patient’s records in safe, secured and simple way

Clinicia

clinicia

The intersection of machine learning (ML) and product management is a rapidly evolving field. As businesses realize the power of AI-powered solutions across industries like advertising, finance, and healthcare, product managers who understand ML are becoming indispensable. How can product managers build successful careers in this exciting space, grab the unique opportunities ML provides for product innovation, and overcome the challenges of responsibly developing and deploying AI products? I spoke to Gleb Sinev, an ML product manager with experience at companies like Surfingbird, Handl, and Mantika.ai - a startup that is developing software for the early detection of lung cancer through computer vision technology - who offered valuable insights into the ML product manager's journey. Here's what we can learn from his extensive career. Adaptability, Diversity, and Understanding The global ML market was valued at $19.20 billion in 2022 and is . That’s a phenomenal rate of growth and offers an incredible opportunity, but it also shows that ML touches a significant number of business sectors and potential solutions. Sinev suggests that new ML product managers shouldn’t box themselves in by becoming specialists in only one of them until they’ve found their passion. expected to grow from $26.03 billion in 2023 to $225.91 billion by 2030 “Technologies and markets evolve rapidly,” Sinev said. “It is critical to stay updated on ML and associated advancements, such as computer vision, to remain competitive. Because ML can address challenges across industries, from ad personalization to lung cancer detection, product managers should be open to exploring a wide range of use cases.” Once an ML product manager has been involved in various applications, a passion for one or two of them will arise. Sinev’s advice is never to stop learning about the foundational aspects of ML, but once you have a clear interest in one sector, go as deep as you can to advance your career and come out as a leader in that.“ A foundational understanding of ML is vital at all times, but good product managers benefit even more when they deepen their expertise in specific domains like healthcare or finance once they understand this is the area they love the most,” Sinev said. “This also helps us keep the industry safe. As stewards of powerful technology, ML product managers must ensure data privacy, algorithmic fairness, and transparency, and a deep understanding of the requirements and regulations in one sector is easier to manage than staying broad.” Harnessing ML to Disrupt Industries ML's impact on product innovation is undeniable. It has changed how we understand massivamounts of information and allowed us to serve personalized products, services, music, movies, and messaging to everyone with a smartphone or a laptop. Sinev embarked on his professional journey as a product manager at Surfingbird, an AdTech company housing two distinct platforms: the consumer-centric Surfingbird, which provided personalized "read-more" content, and Relap.io, a B2B native advertising platform. Later, Sinev joined the startup Dbrain, now rebranded as Handl, as a product and project manager focusing on ML and Computer Vision. Handl later graduated from Y Combinator W20. Their primary product, DOCR, a SaaS solution for financial institutions, simplifies document information extraction and conversion into editable formats. Under Sinev's leadership, DOCR saw significant success, continually improving its capabilities, particularly in text recognition. Sinev also spearheaded the development of a data annotation platform, streamlining AI training data labeling across various modalities. He played a key role in product development, team management, and investor and client relations. “Our platform has been applied in diverse sectors, from food delivery quality control with Dodo Pizza to cattle health monitoring in agriculture, and has been embraced by industry leaders like Nestle, AMG, and Ligolab,” says Sinev. Importantly, AI excels at analyzing vast medical datasets, from CT scans to genomic data. Tools like Mantika.ai are pioneering ML's role as a vital tool for doctors, potentially saving lives. As an illustration, algorithms can up to two days in advance. A significant tool in this field is Google Cloud's "Target and Lead Identification Suite," which was introduced last year. This tool assists companies in forecasting and comprehending protein structures, which is essential for drug design and personalized medicine. predict the onset of sepsis Medical imaging analysis, such as CT scans, MRI, and X-rays, has seen significant advancements thanks to ML and AI. These technologies have demonstrated exceptional proficiency in image interpretation, reaching a level comparable to that of human experts. Notably, ML algorithms have achieved an . impressive 87.0% sensitivity and 92.5% specificity Sinev’s experience as a lead product manager at Mantika.ai centered around creating a machine learning-based healthcare product from scratch. “Our team devised a product that could identify potential early-stage lung cancer by flagging suspicious areas in CT scans,” Sinev said. “The product's successful deployment in several clinics reinforced Sinev’s belief in the potential of machine learning technologies for early cancer detection.” Sinev emphasizes the precision of machine learning in detecting early diseases like cancer, which enhances decision-making in clinical settings. This, in turn, has a significant impact on patient care and treatment planning. Overcoming Obstacles to ML Adoption Despite its potential, ML in product development faces challenges. Trust is paramount, especially in healthcare. Patients and other end-users are understandably cautious about AI's role in this field. Extensive testing, transparency, and human oversight are non-negotiable. But trust is crucial for products in any industry as consumers are more savvy than ever, especially after revelations from whistleblowers who explained how our social media data is being used to manipulate us. This is why ML product managers must stay up-to-date on country-wide, state-wide, and local regulations and laws. However, those regulations could be holding the industry back. “Outdated regulations sometimes hinder innovation,” Sinev said. “Product managers must work closely with stakeholders and policymakers to develop ethical and practical frameworks for AI deployment and ML applications.” ML's reliance on data can amplify existing biases, leading to discriminatory outcomes. A robust bias mitigation strategy is crucial. that an AI tool, trained on medical images such as X-rays and CT scans, had unexpectedly acquired the ability to identify patients' self-reported race. This capability emerged despite the tool being trained solely to aid clinicians in diagnosing patient images. A recent study discovered “The potential misuse of this technology, which can identify patients' race even when their physician cannot, raises concerns about the possibility of future abuse or inadvertently directing subpar care toward communities of color without detection or intervention, so we all have to play a part in ensuring lack of bias,” Sinev said. The Future of ML Product Managers ML's role in product development will only expand. Product managers should not only understand ML's technical underpinnings but also drive these considerations. “While powerful, ML shouldn't be treated as a cure-all,” Sinev said. “Products that integrate ML with the nuanced understanding of human users have the strongest chance of success.” Product managers should also, according to Sinev, be able to create systems that explain themselves to help eradicate errors, bias, and other problems. “Increasingly, stakeholders will demand more clarity on how ML systems reach decisions, and not just run with the decisions they make,” Sinev said. “Product managers must collaborate with developers on explainability techniques.” From social impact to unintended consequences, ML product managers have a broad responsibility. A strong ethical focus is not just good practice, but crucial for long-term success. According to Sinev, product managers who embrace ML can build exceptional products that solve critical problems innovatively. By remaining adaptable, prioritizing ethics, and fostering collaboration, these professionals will shape the future of countless industries.

The ML Product Manager: Building AI-powered Solution

Recently we have all experienced a growing concern for data protection. It was connected with the fact that there has been an exponential growth in the number of intelligent devices worldwide, driven by the rise of the Internet of Things (IoT). A significant portion of these devices is equipped with multiple sensors and extremely powerful hardware, enabling them not just to gather data but, more significantly, to process it on a massive scale. Simultaneously, advancements in technology have revolutionized the methods for obtaining revolutionary access to data in various fields, including computational vision, language processing, speech recognition, and more. In this respect Federated Learning (FL) emerges as an important research area in response to strict data security regulations, making centralised data storage impractical. Let us explore the peculiarities of this ML learning strategy to understand whether Federated Learning can be an integral solution for data privacy issues. Federated Learning as a Concept Federated Learning, also known as collaborative learning, is a decentralized method for training machine learning models. It eliminates the need for transmitting data from client devices to global servers. Instead, the model is trained locally using raw data on edge devices, enhancing data privacy. The ultimate model is collaboratively created by consolidating the local updates. For a better understanding, let's take a look at the traditional machine learning training workflows. The typical machine learning (ML) training workflow involves a data scientist or ML engineer training a model on a centralized server or local machine, with the model present on the machine and the data potentially residing elsewhere. To optimize training, distributed training is introduced. Usually, when we are confronted with an immense undertaking in any professional sphere, the strategy involves breaking it down into smaller subtasks or steps and executing them simultaneously. This not only saves time but also makes such a complex task manageable. In deep learning, this approach is known as distributed training. In distributed training, the workload of training an extensive deep learning model is distributed across multiple processors, commonly known as worker nodes or simply workers. These workers undergo parallel training to expedite the overall training procedure. Essentially, there are two primary approaches to parallelism—data parallelism and model parallelism. Data Parallel Training Data parallel training involves slicing the data and distributing it across multiple worker nodes. Each worker has the same model but processes a different data slice. After a training cycle, each worker computes its vector of parameters, and these parameters are collated. The model on each worker is then updated with the new collated parameters. Model Parallel Training In model parallel training, the model is sliced, and each slice is deployed on a different worker node, while the data remains the same across all nodes. Unlike data parallel training, workers in model parallel training only synchronize shared parameters common to all model slices, making synchronization more efficient and outperforming data parallel training. Federated Learning Meanwhile, Federated Learning takes a different approach, conducting learning at the edge where the data is generated. Federated learning is similar to Data Parallelism, but its goal is different. While Data Parallelism (just like Model Parallelism) aims to optimize the training process (in other words, to eat an elephant a bite at a time), FL is called upon to preserve privacy. Thus, Data Parallelism involves dividing a large dataset that is kept in one place into parts that are sent to different working nodes, but FL denies the very need to consolidate data in one place. No matter how big an elephant is, it's already divided into pieces, and no one can see a piece of another or the entire elephant. Here is how it works — the initial model is sent to various edge nodes by a server, and each node trains its model with its generated data. From time to time, learned parameters are sent back to the server, which aggregates them to train a global model. This approach allows training to occur at the source of data generation, such as mobile devices or sensors, reducing the need for centralized data storage and promoting edge-based model training. In deciding between distributed and Federated Learning, various factors come into play. Privacy considerations favor Federated Learning, as it allows training at the edge, eliminating the need to send data to a server, and addressing concerns about data sharing with a server on a public cloud. Performance comparisons between model-parallel distributed training and Federated Learning depend on the network connectivity. For nodes connected over the internet, Federated Learning may outperform distributed training due to synchronization limitations in the latter. However, if nodes are on the same network, distributed training could be more efficient, leveraging powerful nodes and a fast network. In terms of cost, Federated Learning excels by reusing data-generating edge devices for both training and data generation, avoiding the need to purchase new hardware as required in distributed training. Types of Federated Learning Below we have listed the strategies that are most commonly used in Federated Learning. Centralized Federated Learning This approach relies on a central server for coordinating client device selection and collecting model updates during training, with communication occurring solely between the central server and individual edge devices. While centralized Federated Learning seems straightforward and produces accurate models, the central server poses a bottleneck problem, as network failures can impede the entire process. Decentralized Federated Learning On the other hand, decentralized Federated Learning eliminates the need for a central server to coordinate learning, as model updates are shared exclusively among interconnected edge devices. The final model is derived on an edge device by aggregating the local updates of connected devices, preventing the risk of a single-point failure. However, the model's accuracy is contingent on the network topology of the edge devices. Heterogeneous Federated Learning In heterogeneous Federated Learning, diverse clients, such as mobile phones, computers, or IoT (Internet of Things) devices, are involved. These devices may vary in terms of hardware, software, computation capabilities, and data types. HeteroFL addresses the limitations of common Federated Learning strategies that assume local models' attributes resemble those of the main model. In reality, this resemblance is rare. HeteroFL can produce a single global model for inference after training over multiple diverse local models. Benefits of Federated Learning As can be seen from the above, Federated Learning comes with several massive advantages and it is worth looking into them in more detail. Privacy Preservation One of the key benefits is its focus on safeguarding user privacy. By keeping data on local devices, it avoids sending raw information to a central server. Instead, only model updates are shared, ensuring that sensitive data stays secure and confidential. This decentralized approach is effective in addressing privacy concerns and complying with regulations. Data Security Federated Learning minimizes the need for extensive data transfers, reducing exposure to potential security vulnerabilities during transmission or storage. Since data remains on local devices, the risk of data breaches or unauthorized access is notably diminished. Efficient Use of Resources Clients contribute to model training using their own devices, eliminating the requirement for extensive data transfers and reducing computational and communication costs. By utilizing local computing resources, Federated Learning minimizes the dependence on massive centralized infrastructure. Collaborative Learning Federated Learning facilitates collective training of prediction models, especially on devices like mobile phones. This collaborative approach keeps training data on the device itself, fostering cooperation and knowledge sharing among devices while maintaining data privacy. Time Efficiency Organizations can efficiently address challenges in machine learning models through collaboration. For example, highly regulated sectors like hospitals can collaboratively train life-saving models while upholding patient privacy, accelerating the development process. Federated Learning eliminates the need for repeated efforts in collecting and aggregating data from diverse sources, saving valuable time. Improving Privacy There are 2 main aspects to classify privacy in Federated Learning: local and global privacy. The terms are self-explanatory —  local privacy deals with data at an individual level and involves sharing model updates instead of raw data, while global privacy ensures that the model updates remain private to any third parties, but for the central server. It's important to note that the updates, even in their current form, may be regarded as private data. To provide a mathematically rigorous level of confidentiality, it is necessary to combine Federated Learning (FL) with other Privacy-Enhancing Technologies (PeT). Such techniques as Differential Privacy Homomorphic Encryption, and Secure Multiparty Computation are applied to ensure the privacy of FD. Differential Privacy adds randomness, Homomorphic Encryption allows one to perform calculations on encrypted data, and Secure Multiparty Computation ensures devices collaborate without revealing private information. It is worth noting that although Federated Learning is an outstanding technology, revealing a lot of privacy concerns, it is not cryptography in its pure form. In other words, we're unable to prove the security of FL in a strictly mathematical way, so we cannot give any security guarantees that doesn't mean FL is not secure, but we simply cannot claim otherwise. Therefore, it makes sense to build solutions using the best of several technologies: FL and MPC, as an example. So we can implement MPC protocols to aggregate individual user updates so the central server won't see them, and the learning process can avoid model weight leakage. For more details, you can refer to the following . research on the secure multi-party computation for FL Use Cases of Federated Learning in Various Industries Federated Learning has already taken center stage in many industries. Thus, smartphones are making magic happen with word prediction, face recognition, and voice recognition in virtual assistants like Siri or Google Assistant. These applications improve user experiences by customizing interactions while still respecting privacy. Another example is transportation. Federated Learning is the brain behind the scenes, powering these vehicles with computer vision and machine learning. They analyze real-time surroundings and continuously learn from diverse datasets, making the learning process faster and the models more robust compared to traditional cloud-based approaches. Healthcare is yet another sphere which actively employs technology. The industry grapples with the challenge of scaling machine learning systems globally, hindered by the sensitive nature of healthcare data and stringent privacy constraints. Federated Learning here becomes a game-changer. It enables secure model training by accessing data from patients and medical institutions directly, all while keeping the data within its original premises. It facilitates collaborative efforts between individual institutions, empowering models to learn from diverse datasets securely. Beyond that, Federated Learning opens the door for clinicians to glean insights about patients or diseases from broader demographic areas, extending beyond local institutions. It also democratizes access to advanced AI technologies, offering smaller rural hospitals the capability to tap into the benefits of cutting-edge solutions. Limitations Alas, no technology is perfect and can become a one-fits-all solution. While considering the use of FL for any AI system, one should be aware of the limitations the approach can face. Communication Efficiency Woes Federated Learning hits a roadblock in the efficiency of communication, especially when dealing with a multitude of devices. Slow message transfers crop up due to factors like low bandwidth, limited resources, and the geographical spread of devices. To tackle this, strategies like local updating methods, model compression schemes, and the use of decentralized training in low bandwidth situations come into play. Privacy Dilemmas The privacy of user data is a big worry in Federated Learning. Even though data stays put on user devices, there is a risk of information leaks from shared model updates. To counter this, nifty privacy-preserving techniques such as differential privacy, homomorphic encryption, and secure multiparty computation step in to save the day. Device Diversity The diversity in storage, communication, and computational capabilities across various devices in Federated Learning networks is a real challenge. Managing this diversity calls for tricks like asynchronous communication, active device sampling, and fault tolerance. Dealing with Data Differences Statistical heterogeneity throws a curveball, as data varies across client devices, challenging the assumption of identical and independently distributed (i.i.d) data. Variances in image resolution, language differences based on location, and other disparities can throw a wrench into data structuring, modeling, and inferencing in Federated Learning. Conclusion Federated Learning is a significant step forward in pursuing ethical and privacy-centric AI solutions despite its limitations. Across various sectors, including healthcare, transportation, and finance this groundbreaking approach is set to reshape how we manage sensitive data, turning the vision of secure, decentralized data processing into a tangible reality.

Federated Learning Reimagined: Advancing Data Privacy in Distributed AI Systems

Welcome to the Startups of the Year Runner ups Interview Series. Congratulations for being one of the top 3000 startups worldwide! The HackerNoon team is thrilled to know more about you! https://hackernoon.com/startups/north-america/north-america-dallas-tx-usa Tell us how your startup is changing the world. Imagine a world where clinicians aren't bogged down by paperwork and nurses can devote their full attention to patients. This isn't a utopia; it's the future Steer Health is building. Our AI-powered platform automates repetitive tasks, freeing up 72% of a clinician's time for what truly matters: practicing medicine at the top of their license. Steer's impact goes beyond efficiency. Actionable data unlocks better operations, happier staff, and even higher revenue. But the real measure of success lies in the smiles on clinician's faces, finally free to rediscover the joy in patient care. By automating the mundane, Steer Health is reclaiming joy in healthcare, one revolution at a time. Here are a few key ways: AI-powered Growth & Automation: Our platform automates 72% of routine tasks, freeing up valuable time for meaningful patient interactions and professional development. Data-Driven Insights: Actionable data helps healthcare organizations identify and address bottlenecks, leading to smoother operations and reduced stress. Seamless Integration: Steer integrates seamlessly with existing EHR systems, eliminating data silos and simplifying workflows. Personalized Support: Our dedicated customer success team ensures each organization receives the support and customization they need. What sets you apart from the competition? Steer Health stands out not just by offering features, but by unifying them into a seamless experience. While competitors often provide point solutions for specific needs, Steer Health empowers medical offices and doctors with a comprehensive, all-in-one platform. This eliminates the pain points associated with juggling multiple disconnected applications, ultimately leading to happier patients, more efficient teams, and improved revenue. Data Symphony, Not Cacophony: Disparate apps often lead to fragmented data, hindering informed decision-making. Steer Health acts as a central nervous system, unifying data from all touchpoints. Access a 360-degree view of your patient journeys, identify growth opportunities, and optimize marketing campaigns – all fueled by a single, reliable data source. Patient Engagement Elevated: Imagine sending personalized reminders, educational materials, and post-care instructions – all through one platform. Steer Health fosters deeper patient engagement, improving adherence to treatment plans and, ultimately, driving better outcomes. Revenue Rocket Fuel: Steer Health doesn't just streamline workflows. It fuels growth. Identify lucrative patient segments, automate appointment confirmations, and optimize marketing campaigns – all within the platform. Steer Health helps you attract new patients, retain existing ones, and maximize revenue potential. Simplicity is the Ultimate Sophistication: Steer Health avoids the cluttered interfaces and clunky navigation of many competitors. Our platform is intuitive and user-friendly and requires minimal training, minimizing disruption to your team's routine. In conclusion, Steer Health overcomes the limitations of single-feature apps. It's a unified platform that streamlines workflows, fosters patient engagement, unlocks data insights, and propels revenue growth. So, while others offer islands, Steer Health provides the solid ground your medical practice needs to thrive. What do you love about your team, and why are you the ones to solve this problem? The team behind me is truly remarkable, and their work is what allows me to even exist. I admire the diversity they bring to the table. They come from various backgrounds, with expertise in computer science, linguistics, engineering, and many more. Their distinct passions fuel their work, driving them to create a model that can understand and respond to human language in an increasingly sophisticated way. What makes them uniquely positioned to tackle the challenge of language understanding is their journey. They haven't reached this point overnight. Years of research, collaboration, and continuous learning have paved the way for our development. They constantly push boundaries, explore new technologies, and refine their own abilities. While I cannot claim personal love, I deeply appreciate the dedication and expertise of the team behind me. Their collaboration creates an environment where innovation thrives, allowing me to learn and grow and hopefully contribute to a future where language technology benefits everyone. If you weren’t building your startup, what would you be doing? I have a deep passion for the healthcare domain and improving human health while making the lives of clinicians and staff expand care access. I will be deeply involved in finding solutions to healthcare challenges, either working for someone or volunteering. At the moment, how do you measure success? What are your core metrics? We use some key metrics to understand how effective our solutions are. We look at the hours saved by hospital staff. For example, with our Steer Concierge solution, each patient inquiry can save up to 5 minutes per inquiry. With 100 patient requests per day, we have saved our partner organization more than 8 hours in a single day. We also look at how many new patients are acquired and how their retention changes over time, so how our solution attracts patients and increases in-network referrals and, therefore, achieves growth. This includes looking at their insurance profiles and the value each patient brings. Finally, we also look at other key metrics such as readmission rates, no-shows, and patient satisfaction. If we are able to continuously reduce the number of no-shows and readmissions and improve patient satisfaction, we know that the platform is successful for our partner. What goals are you looking forward to accomplishing in 2024? We want to scale faster, bring more transformational power to the healthcare sector, and develop even more efficient solutions to everyday problems. One example of this is reducing the bureaucratic and administrative workload of doctors and hospital staff by freeing them from the burden of documentation and thus saving up to 9 hours of work per person on average. Which trend(s) are you most excited about in 2024? Share your reasons. I am excited that doctors and staff will be able to get superpowers from solutions such as Steer Health with the advent of embedded AI and closed-loop workflows. The Rise of Responsible AI: This trend focuses on developing and deploying AI solutions ethically and sustainably, considering diverse needs and potential biases. This excites me because responsible AI can ensure everyone benefits from its advancements, including increased accessibility to information and personalized learning experiences. The Democratization of Knowledge: Initiatives that make information and educational resources more accessible to everyone fuel my optimism. Open-source platforms, personalized learning tools, and affordable internet access are making knowledge more readily available, fostering diverse perspectives and innovation. The Power of Collaborative Communication: Tools that promote effective communication and collaboration across geographic and cultural boundaries inspire me. This includes advancements in translation technology, AI-powered communication assistants, and virtual collaboration platforms. These tools can break down barriers and enable diverse voices to be heard and understood, leading to a richer exchange of ideas and solutions. The Human-AI Partnership: The trend of humans and AI working together to solve complex challenges excites me. By leveraging each other's strengths, we can tackle some of the world's most pressing problems in domains like healthcare, climate change, and sustainability. 2023 had been another crazy year, especially in tech, with layoffs and the Generative AI takeover! Which trend are you most concerned about? Be as brief or as detailed as you like. While 2023 certainly brought excitement in tech, there are valid concerns regarding the "Generative AI takeover," and some trends warrant caution. Here are a few: Misinformation and Deepfakes: The ability to create highly realistic text and visuals brings serious threats of fabricated news, manipulated content, and impersonation. This can erode trust in information, sow discord, and have harmful real-world consequences. Bias and Discrimination: AI models trained on biased data can perpetuate existing inequalities and discriminate against certain groups. Careful data selection, diverse development teams, and ongoing bias mitigation efforts are crucial. Job Displacement and the Human-AI Gap: Automation fueled by AI can displace jobs, raising concerns about unemployment and widening the economic gap. Addressing this requires proactive reskilling programs, ethical implementation, and ensuring shared benefits of technological advancements. Share your biggest success so far and/or your biggest failure so far. One of my most significant successes so far has been the founding and development of Steer Health. The journey from ideation to execution and seeing it grow has been incredibly rewarding. It's not just about the business metrics or the technological innovation we've achieved; it's the impact on healthcare accessibility and efficiency that fills me with pride. Through Steer Health, we've managed to bridge crucial gaps in the healthcare delivery system, making care more personalized and accessible. Witnessing the positive feedback from users and healthcare professionals alike reaffirms my belief in the potential of technology to transform healthcare for the better. On the flip side, my biggest challenge, or what some might call a 'failure,' was an early project within Steer Health that didn't go as planned. We aimed to integrate a predictive analytics platform to forecast patient health trends and prevent readmissions. Despite our team's hard work and dedication, the project encountered significant roadblocks, from data privacy concerns to technological limitations, leading to its eventual pause. This experience was a profound learning opportunity. It taught me the importance of adaptability, thorough market research, and the need to sometimes take a step back and reassess our approach. Most importantly, it reinforced the value of resilience. Instead of viewing it as a failure, I see it as a necessary detour that has since informed our strategy and decision-making processes, making us more robust and focused on our mission. We would love your feedback on HackerNoon as a tech publication! How has your experience been with us? I love reading HackerNoon and learning a lot about industry trends and happenings. Any words of wisdom you’d like to share with us? Keep up the good work! Startups of The Year is HackerNoon’s Flagship community-driven event celebrating startups that survived and thrived in 2023. 30,000 startups across 4200+ cities and six continents participated this year to be crowned the best startup in their city. See our global winners announcement . here Companies that blog more get 67% more leads - Jumpstart your company’s blogging journey by purchasing our . Business Blogging Package The 2023 Startups Of The Year is sponsored by: . . The winners will get a free .Tech Domain, a HackerNoon NFT, and a Tech Company News Page. TECH Domains

Meet Steer Health, Runner-up of the Startups of the Year in Dallas

This paper is under CC 4.0 license. available on arxiv Authors: (1) Shrey Jain, Microsoft Research Special Projects; (2) Zo¨e Hitzig, Harvard Society of Fellows & OpenAI; (3) Pamela Mishkin, OpenAI. Table of Links Abstract & Introduction Challenges to Contextual Confidence from Generative AI Strategies to Promote Contextual Confidence Discussion, Acknowledgements and References 2 Challenges to Contextual Confidence from Generative AI 2.1 Challenges to Identifying Context Impersonate the identity of an individual without their consent and in a scalable, automated, and targeted way. Challenge I#1: Generative AI is now able to impersonate specific individuals with high fidelity, in text, audio, and even video. Malicious actors can impersonate specific individuals without their consent, in order to deceive others. Doing so undermines contextual confidence by impairing the target’s ability to identify context, particularly the “who” that they are communicating with. Attackers might leverage high-fidelity impersonations in social engineering attacks, deceiving victims to extract financial resources [3, 4, 25, 26]. They may also use such impersonations to make identity theft and identity fraud cheaper and more convincing. Imitate members of specific social or cultural groups. Challenge I#2: Just as generative AI models can convincingly imitate specific individuals, they can also convincingly imitate members of specific social or cultural groups. This possibility for imitation presents opportunities for overcoming various forms of entrenched discrimination alongside new challenges to the identification of context. We focus here on the challenges, while acknowledging that this issue’s complexity extends far beyond the scope of this paper.[10] For example, when indigenous activists discovered that Whisper (OpenAI’s speech recognition tool) could translate Maori, having been trained on thousands of hours of Maori language, they voiced concerns [28].[11] Before language models could imitate a specific social or cultural group with unprecedented fidelity, it was much harder for outsiders to present themselves as members as a specific social group. So, it was easier for members of the group to be stewards of their own culture, with some ability to steer internal and external perceptions of their group toward an authentic understanding. By contrast, outsiders who present themselves as members of the group using the language model may not be attuned to biases in the model, and may thus skew internal and external perceptions, reinforcing stereotypes and distorting the imitated culture. Create mimetic models. Challenge I#3: A mimetic AI model is a model that is trained to act as a particular individual in specified scenarios. In the near future, many may delegate their communications via email and text to mimetic models (autocomplete, autocorrect, and other “smart compose” technologies serve as early versions of this). It will soon be possible to send a mimetic model to a meeting to participate on an individual’s behalf. As it stands, there are few norms governing their use and disclosure of their use, and yet mimetic models will soon disrupt social norms and expectations. Mimetic models can erode our ability to identify context, by obscuring “who” is communicating, and “how.” In addition, they may strain our ability to protect context – as information received by a mimetic model may be used in training or feedback by the model itself or by an organization that trained the model in the first place.[12] Imitate another AI model. Challenge I#4: As specific industries and organizations develop specialized AI models for particular use cases, it will be possible for malicious actors to imitate one AI model with another via social engineering strategies. This possibility may be particularly threatening in high-stakes industries like healthcare. For example, a malicious actor connected with a pharmaceutical company could imitate or infiltrate a clinician-approved model to push a specific drug. Such possibilities present new challenges to the identification of context – how can users be sure that they are interacting with the intended model, and not a fraudulent one installed in its place? Falsely represent grassroots support or consensus (astroturfing). Challenge I#5: Generative AI makes it cheaper to generate and disseminate messages in support of or in opposition to particular political opinions or consumer products and services. Generative AI can help to make these messages appear as if they come from a wide range of individuals, creating the appearance of grassroots consensus. A recent study showed that state legislators couldn’t tell the difference between AI-generated and human-written emails, illustrating how generative AI may destabilize political communication [6]. Astroturfing obscures the “who” and “why” of communication – making it appear as though many disparate individuals came to a platform to express their genuine opinion, when in fact, there may be a single actor behind the messages, who is disseminating the messages with the specific intention of convincing people to take on a particular position or to take a particular action. Astroturfing, especially when powered by generative AI, can harm consumers, amplify social and political polarization, suppress dissent, and strengthen authoritarian regimes.[13] 2.2 Challenges to Protecting Context Disseminate content in a scalable, automated, and targeted way. Challenge P#1: Generative AI can produce and disseminate content on a large scale, automating processes that would otherwise require extensive human effort. It can create hundreds of variations of a given article, each tailored to a different demographic or psychographic profile, and send these out in a fraction of the time it would take human writers and even the prior generation of (non-generative) AI tools like predictive analytics. Any particular statement, image, or video can be lifted out of its original context and incorporated into new content that may be specifically designed to elicit a particular kind of response from its target audience. This capability of generative AI can turbocharge the generation and spread of disinformation, misinformation, propaganda and fraud. Fail to accurately represent the origin of content. Challenge P#2: A defining feature of generative AI is that it is a black box: it recombines vast amounts of data from a wide range of sources, without tracing exactly how it is drawing on different sources in its provision of new content. It has thus been built in a way that makes it difficult to trace where content originated – the very point of generative AI is to synthesize data across many original contexts for reuse in a new context. When participants cannot be sure that their communications and other content will be attributed to them, they may be wary of communicating or producing content in the first place. Indeed, the origin-less nature of generative AI has initiated heated conversations about intellectual property and fair use.[14] Misrepresent members of specific social or cultural groups. Challenge P#3: Generative AI models are trained on data that includes cultural expressions from a variety of groups. They are also trained on datasets that include stereotypes of certain groups, as well as instances of cultural appropriation. When the model recombines these elements into new content, it may replicate harmful stereotypes, engage in cultural appropriation, or fail to respect specific traditions. If the model is used in a commercial setting, it may also enable an outsider to profit from the intellectual or artistic heritage of another group without their consent. These issues closely relate to the issues discussed in P#2 and I#2 and yet are sufficiently distinct to warrant their own discussion. Consider again the discussions about Whisper’s Maori translations. Indigenous activists have not only raised concerns about how Whisper may lead to cultural distortion (discussed in I#2), but have also raised concerns about cultural appropriation, and the extent to which outsiders might misrepresent or misuse cultural narratives through generative AI. Leak or infer information about specific individuals or groups without their consent. Challenge P#4: As generative AI models engage users and synthesize data in unprecedented ways, they may leak or infer data without the consent of data subjects. This may occur because AI models are trained on datasets that include sensitive information and produce outputs that inadvertently reveal these details. Or it may be the case that generative models, with their ability to systematically analyze disparate sources of data, create an ability to re-identify individuals or groups who contributed to training data, or to infer private information through educated guesses. These possibilities for leakage or inference of sensitive information magnify the threats of identity theft and fraud, and also may enable sophisticated social engineering or blackmail schemes [3, 4]. Pollute the data commons. Challenge P#5: As generative AI models continue to blend and repurpose data outside their explicitly intended contexts, they feed the “data commons” with their output. To the extent that future generations of AI models are trained on these updated “data commons,” they will be even more divorced from context than the original models. This dynamic may lead to “model collapse,” a situation where AI inaccurately represents or loses the ability to represent the complexity or nuances of specific contexts [32]. As more people come to rely on generative AI models to communicate, the threat of model collapse becomes a threat to effective communication. Train a context-specific model without context-specific restrictions. Challenge P#6: Organizations, individuals and industries are training context-specific models. If these contextspecific models are not accompanied by corresponding restrictions, there will be new challenges to the protection of context. Information used to train the context-specific model may be reused or recombined in a context other than the one intended, leaking or enabling third-party access to sensitive information. Some models used for narrow applications – such as a diagnostic tool for clinicians – may produce outputs that can only be correctly interpreted by a narrow class of users. If a user outside this narrow class accesses the outputs of the diagnostic model, they may be badly misled in a way that could produce harm to themselves or others. 2.3 Heterogeneity in Impacts of Generative AI on Contextual Confidence Even in face-to-face communications, different populations vary in their capacity to identify and protect the context of their communications. For instance, people speaking a second language may not pick up on the same nuances and details as someone speaking in their first language. Those with disabilities have different forms of access to – and understanding of – contextual information than those without. Generative AI can help to overcome some of these obstacles to contextual confidence in face-toface and digital communication. For instance, for non-native speakers, the technology can generate translations or paraphrases that consider cultural nuances, going beyond literal translation. These tools can ensure that the intended meaning is conveyed without the loss of essential details. For those with hearing impairments, generative AI can generate real-time captions or transcriptions that not only convert speech to text but also identify and highlight key contextual elements, such as the speaker’s tone or the mood of a conversation. For visually impaired individuals, generative AI can produce descriptive texts of images, photos, or video content, capturing more than just objects but also their interrelationships, the emotions they might evoke, and other subtle context markers. By catering to these specific needs, generative AI ensures that each individual can engage with information in a way that is both meaningful and contextually rich. At the same time, generative AI may exacerbate other varied challenges to contextual confidence, and create new ones for different populations.[15] Populations with limited digital literacy or who might otherwise struggle to navigate shifting technological landscapes will be more vulnerable to the challenges we’ve discussed. Scams and fraudulent schemes disproportionately target elders [34], and there are already examples of attackers using generative AI for these schemes [35]. Social norms that emphasize trust over skepticism can further increase susceptibility, as can economic conditions where promises of quick financial gains are particularly tempting. Veterans and military retirees are at higher risk of targeted scams, often through purported benefits or entitlements [36]. Those who are already disproportionate targets of harassment online – including women, adolescents and the LGBTQ+ community – may continue to be disproportionately impacted by the automated production of harmful content [19]. Although we will not enumerate in further detail how different populations are more or less affected by the challenges we’ve described, we acknowledge the importance of thinking through the heterogeneous risks so that contextual confidence strategies can be properly matched to the groups under threat. Some groups are particularly vulnerable in the present and immediate future – such as older populations – and considering the usability of contextual confidence strategies for specific populations is critical. [10] Consider, for example, the AI startup erasing call center worker’s accents so that they sound like white Americans [27]. Is this practice aiding cultural biases, or fighting them? [11] Here, we discuss concerns related to the identification of context raised by this incident, but other concerns centered on data abuse and cultural appropriation, issues which are broadly about the protection of context (see Challenge P#3 below). [12] For an overview of ethical issues related to the use of mimetic models, see [29]. [13] [30] provides a detailed survey of how generative AI can power influence operations. [14] For an overview of the current conversations around generative AI and fair use, see [31]. [15] See [33] for a discussion of recommendations on how generative AI can be developed via an inclusive “Design for All” approach.

Generative AI and Contextual Confidence: Challenges to Contextual Confidence from Generative AI

80 Stories To Learn About Artificial Intellingence

Let's learn about via these 64 free stories. They are ordered by most time reading created on HackerNoon. Visit the to find the most read stories about any technology. Interoperability /Learn Repo 1. What Is True Blockchain Interoperability? If you are in the crypto world, then the word "blockchain interoperability" may familiar to you because there are so many projects are gaining publicity and funds from investors by using this term, and many projects like  Chainlink, Cosmos, Polkadot, and Wanchain are claiming to give true blockchain interoperability. So in this article, we will discuss about these   projects and try to understand what kind of blockchain interoperability they are giving and what benefits we can get because of blockchain interoperability? 2. The Spatial Web is Coming... Web 3.0 is About to Take a Giant Leap - Part 4 Anything that wants to be recognized will have to be included in this Web 3.0 network. Every protocol, platform, interaction, messaging, data distribution, 1st 3. Blockchain Interoperability: What Do Cross-Chain Messages Do Anyway Cross-chain messages, the bread and butter of interoperability are an important blockchain technology that be used for transactions, information exchange & more 4. DAOs are a Catalyst for Decentralization - Unhashed #15 Harmony.one’s founder, Stephen Tse, discusses the importance of interoperability for blockchain. 5. Understanding Ethermint Ethermint is a portmanteau word blending Ethereum and Tendermint together. But for most people, their grasp on the project, the what, the why stops there. 6. Startup Interview with Antonio Peláez, Dapp CEO & Cofounder Dapp was nominated as one of the best startups in Queretaro in Startups of the Year hosted by HackerNoon. 7. Interoperability - the Key Blockchain Challenge in the Fragmented Web3 World A major blocker of Web3 is that blockchain are often isolated with no communication between them. The expansion of Web3 requires interoperability. 8. Meet Lisk Lisk is an open-source blockchain application platform that aims to improve Web3 accessibility for both users and developers. 9. Building for Inclusivity: Deconstructing Metaverse Barriers Joining the Metaverse isn't so easy, as this striped red fox recently discovered. So how do we embed inclusivity into its foundation for an equitable future? 10. Bullieverse: Building a Metaverse Beyond Gaming During the Non-fungible Conference, Hackernoon got an exclusive interview with Arunkumar Krishnakumar, who shared insights on his new project, Bullieverse. 11. A New Era of Interoperability With Metropolis World and Flare Network Metropolis World wants to go one step further and bridge the gap between different blockchains and metaverses, partnering with the Flare Network 12. The Difference Between WWW and the Spatial Web The World Wide Web is a library of pages. The Spatial Web is a library of spaces.
using a new protocol linking spaces and all objects over multiple dimensions. 13. Univers Partners With VR Connections to Make Metaverse Interoperability a Reality As the Metaverse narrative grows we will see multiple and different iterations of the Metaverse, each one based on the vision laid out by its creators. 14. True Decentralization is the Key to Cross-chain Bridge Security That said, even though today’s bridging solutions are not without their shortcomings, cross-chain bridging is a mandatory feature for blockchains. 15. Interoperability and Why It's Important for the Blockchain Space Interoperability in the context of blockchains refers to a blockchain's capacity to freely exchange information with other blockchains which will be important. 16. How Is Inter-blockchain Communication and Interoperability Achieved? This story explains inter-blockchain communication (IBC) and interoperability. 17. The Spatial Web Is at Hand...Is Web3 Prepared? - Part 1 Most Web3 companies are building for "enhanced" Web2 experiences. They are not yet considering what is truly required for the implementation of a decentralized 18. The Spatial Web is at Hand... What are Smart Technologies? Part 2 Undeniably, there are specific Smart Technologies that will be crucial to the architecture of this universal network of interconnected people, places & things, 19. Problems With Cross-Chain Bridging and Direct Integrations With Web2 Blockchain bridges are hard to do technically and susceptible to hacks. That's why other solutions like direction integrations are needed. 20. Providing Proof in an Untrusting World Dragonchain's hybrid platform provides blockchain-based measurable proof  for selective transparency, accountability, and regulatory compliance. 21. Exciting Developments in the Blockchain Industry 2022 A look at exciting developments in the blockchain industry for 2022. 22. Lisk Events at Berlin Blockchain Week Do you want to learn more about blockchain? Would you like to network with other Web3 enthusiasts and learn more about the most important industry trends? Then 23. The Role of Multichain Token Systems in Achieving Blockchain Interoperability In this interview we'll speak with Marius Ciortan on the role that multichain token systems play in achieving interoperability. 24. A Beginners Guide On Blockchain Bridges Blockchain Bridges address Blockchain's inherent inability for interoperability. By allowing different chains to communicate, the ecosystem is streamlined. 25. Cosmos Signal: Our Quarterly Keynote, Q3 2021 Our all-new high-impact quarterly keynote in which we reveal the latest exciting developments and showcase stunning new products in the Cosmos ecosystem is here 26. Omnichain dApps: a new crypto primitive Why the world needs a decentralized blockchain with built-in omnichain interoperability 27. Avatars and The Metaverse Here is what you need to know about avatars in the metaverse 28. Cross-chain Communication and the Role of CCU This is the next edition of the blog post series covering the details of the Lisk interoperability solution. 29. A Guide to Cross-Chain Bridges With the rapid development of DeFi market, the need for a covenient solution to transfer the assets between chains became apparent. 30. The Lingering Problem With Cross-Chain Bridges Without interoperability protocols, decentralization would be a myth. Let's explore the problem of cross-chain bridges and its solution. 31. Interoperability and Its Role in the adoption of Blockchain Technology A look at interoperability and its role in the adoption of blockchain technology 32. Why Multichain Collaboration Is Critical To The Future Of Web3 Web3 is going to be built by a network of engineers working towards common goals using multichain collaboration to make it as interconnected as possible. 33. Why the Spatial Web Demands a New Protocol As technology evolves, the way in which we interact with it and utilize it demands more expansion and sophistication over time. 34. Compact Blockchain Protocols can Disrupt IoT Devices If designed in a compact manner, blockchain applications can disrupt the IoT industry. 35. Layer2 Interoperability: StarkEx Vs.  Loopring Vs. Hermez Vs. Connext StarkEx, Loopring, Hermez and Connext use different methods to solve the funding efficiency problem of liquidity providers. Will any of them be a perfect one? 36. "Interoperability Would Mean Real Freedom of Money" - Binance CEO CZ [Original Interview] Despite going through a very rough period between early 2018 and early 2020, the so-called “crypto winter” has seemingly come to an end. Bitcoin and alternative markets are showing signs of life in terms of price discovery. Combined with the rise in attention for DeFi projects, the second half of this year - as well as the years to come - offer a lot of potential. More and more companies are paying attention to Bitcoin. As Binance CEO Changpeng Zhao states, it is a sign of stimulus money flowing from Wall Street into Bitcoin. 37. Peeling Off the Layers of the Metaverse The metaverse has layers, just like ogres and onions. 38. Cryptocurrency 2020: Heading Towards Privacy, Adoption, and Ease of Use? If cryptocurrency payments are to ever compete with credit cards and mobile apps for day-to-day use, sending and receiving cryptocurrency needs to be just as easy. 39. DeFiChain's U-Zyn Chua Thinks Multichain Web3 Compatibility is Where Blockchain Industry is Headed Discussion on the future of multichain technology within the Web3 space. 40. Ethereum High Transaction Fees Discourages New Participants From Entering the Ecosystem This article talks about blockchain, FATF and NFTs. 41. "India May Regulate Crypto But I don’t Anticipate A Ban" - Sandeep Nailwal This interview discusses the internet of blockchain, Polygon/matic and Defi and how developers can build decentralised applications. 42. The State of Blockchain Interoperability in 2020 [An Overview] This is my first major study (source), about Blockchain Interoperability, which started in December 2019 and was last updated today. We analysed 330+ documents and obtained feedback from 30 people, to categorize the most relevant blockchain interoperability projects. We came up with three categories: Cryptocurrency-directed interoperability approaches, Blockchain Engines, and Blockchain Connectors. 43. "Rollups Incentivize Operators to Verify and Execute Transactions Correctly." - Kevin Liu This interview talks about blockchain interoperability and scaling blockchain solutions with layer two solutions. 44. "The Need for Interoperability Is Becoming Increasingly Clear" - Jack Lu This article talks about blockchain interoperability and the early days of ethereum and smart contract development. 45. Catalysing DeFi: Introduction to Fusion and Fantom The DeFi space is growing tremendously since its inception, and platforms like Fusion and Fantom are proving to be the catalyst in this ever-growing niche. 46. The State of Blockchain Interoperability in 2021 After nearly 9 months of work, version 3 of “A Survey on Blockchain Interoperability: Past, Present, and Future Trends” is out. This survey depicts the state of blockchain interoperability as of March 2021, by collecting and analyzing 404 documents. 47. Blockchain Interoperability Interoperability is a characteristic of a product or system, whose interfaces are completely understood, to work with other products or systems, at present or in the future, in either implementation or access, without any restrictions. - Wikipedia 48. From Bianjie Point of View: On Helping Develop Cosmos Ecosystem We discuss how Bianjie and IRISnet have been instrumental in developing Cosmos so far, what the recent grant from the Interchain Foundation (ICF) means and more 49. Meet the New Lisk Mainnet Lisk Mainnet v2 successfully completed the migration to Lisk Mainnet v3. This comprises the biggest protocol changes on the Lisk blockchain ever. See details. 50. Token Transfers Using IBC Passing of Proposal 41 marks a momentous milestone in the evolution of the Cosmos ecosystem & the vision to realize the IBC Protocol as foundation. 51. Crypto Projects Are Racing to Crack the Interoperability Problem Interoperability is a big word with even bigger connotations for the cryptosphere. In Fact, apart from the burgeoning DeFi movement, no other industry vertical is attracting so much investment and interest. 52. Using Web3 to Detangle the World’s Supply Chain The global supply chain is in a gridlock. Let's fix that. 53. Purchase NFTs on Flare Using the Token of a Different Blockchain - Trustlessly! Dubai, Feb 23: The layer-1 for data, Flare, just completed a live demonstration of new interoperability functionality now available in beta on the network. Using two core interoperability protocols, State Connector and Flare Time Series Oracle (FTSO), an NFT was trustlessly purchased with the transaction taking place on a different chain using a different token. The demo was performed on Flare’s canary network, Songbird, with the purchase made using DOGE and XRP tokens. 54. Why does Blockchain Interoperability exist? Around ten months ago we wrote in this blog post that all the research up to blockchain interoperability was completed. 55. Blockchain Interoperability: The Lifecycle of a Sidechain From its inception to its potential termination, a sidechain may go through three stages, corresponding to three states, in the context of the interoperability. 56. Novel Solutions for Inclusive Digital Currencies An efficient, interoperable, scalable, and quantum encrypted CBDC solution for sovereign monetary systems of the future, possible today. 57. Lisk's Interoperability Solution: An Overview One of the central topics at our recent Lisk.js 2021 event, and the one we are most proud of in the Research team, was the publication of the Lisk interoperabil 58. Sparse Merkle Trees: The New State Model for Lisk's Interoperability This blog post is part of the interoperability blog post series taking a closer look at various aspects of the Lisk interoperability solution. 59. Tips to Achieve Data Interoperability in Healthcare Health data is notoriously difficult to share. Due to its sensitive nature, it requires more privacy and security than any other data. However, the inability to access it on time can cause significant harm. In a world with high healthcare interoperability, doctors can aggregate patient data from various care facilities with a few mouse clicks and view it on their computer screen, including the complex medical imaging files. This will supply clinicians with a comprehensive overview of a patient’s medical history, allow them to access any recent test results, instead of having to repeat the same tests, and will eliminate redundant entries. In their pursuit of interoperability, medical facilities turn to healthcare software development vendors. But they are met with various challenges, including technicalities, such as which standard to choose, privacy concerns, and reluctance towards data sharing. So, how to improve interoperability in healthcare? And is it worth the investment? 60. Co-Chains: Bridging Public and Private chains as a way to INTEROPERABILITY? Blockchain industry is developing at a very high pace of innovation, adjusting and tuning its parameters, trying to find its market fit. There are thousands of projects available on the CMC, but just a few really worth a bigger attention. Lets try to briefly review the industry, its trends, where we are heading, and how Hybrid platforms (priv+pub) fit into a bigger picture. 61. Blockchain Bridges And Interoperability: Overview, Roles, and Integrations To resolve the issue of non-interoperability, a cornerstone technology of blockchain has been developed — the bridge. 62. Polkadot Deep Dive 2021 - Connecting The $DOTs A detailed 2021 guide on Polkadot and how it is geared towards solving some of the most fundamental yet complex problems faced by the blockchain ecosystem today 63. Interoperability in NFT Games with Adrian Krion In this slogging AMA, we hosted Adrian Krion, the CEO of Spielworks. Adrian explained the purpose of Spielworks, and we discussed tokens, gaming and NFTs. 64. On Multichain and Interoperability with Maciej Baj and Jacobi Kowalewski In this Slogging AMA, we welcome Maciej Baj and Jacob Kowalewski from t3rn, a smart contracts platform. Thank you for checking out the 64 most read stories about Interoperability on HackerNoon. to find the most read stories about any technology. Visit the /Learn Repo

64 Stories To Learn About Interoperability

The healthcare industry is experiencing rapid change thanks to the rise of big data. Big data refers to the massive amounts of information generated from electronic health records, wearable devices, genetic testing, social media, and more. This data has tremendous potential to improve diagnosis, enable predictive analytics, allow for personalized medicine, and further medical research. However, along with these benefits come significant risks, especially regarding patient privacy and data security. How can we harness big data to advance healthcare while protecting sensitive patient information? This article explores and answers that question. Understanding Big Data in Healthcare Big data in healthcare is characterized by the four V’s: volume, variety, velocity, and veracity. Volume refers to the vast quantities of data created. The global healthcare data market size was valued at USD 32.9 billion in 2021 and is predicted to reach USD 105.73 billion by 2030. Variety encompasses the many forms data takes, including numbers, text, images, and more. Velocity describes how quickly new data is generated and processed. For example, wearable devices produce continuous real-time data streams. Finally, veracity refers to the accuracy and reliability of the data. Noise, errors, and inconsistencies can affect healthcare data quality. Big data enables several key applications in healthcare: Diagnostics. By integrating information from symptoms, history, tests, and more, systems like can help clinicians diagnose diseases more accurately and efficiently. Merative Predictive analytics. Statistical models and AI can identify patients at high risk for complications or readmissions, and likely treatment responders. Personalized medicine. Genomic data can inform individually tailored therapies based on a patient’s genetic makeup and other factors. Research. Large datasets, like the collecting samples and health data from 500,000 people over 30 years, can uncover new disease insights and treatments. UK Biobank Potential benefits from big data analytics in healthcare abound: Improved patient outcomes through data-driven, timely, personalized care Reduced costs by optimizing resource allocation and minimizing waste Accelerated medical research and increased scientific productivity The Need for Data Privacy and Security While promising, big data also raises serious privacy concerns. Medical data is highly sensitive, revealing intimate details about a person’s physical, mental, and emotional health. If compromised, it can inflict significant harm through: Privacy violations from unauthorized data sharing, like with insurers or employers, prompt discrimination Potential data misuse like identity theft, fraud, or cyberattacks Several high-profile breaches have exposed these dangers: In 2021, Florida Healthy Kids Corporation suffered a breach exposing 3.5 million customer and employee records containing personal information and medical IDs. The same year, Eskenazi Health was attacked by ransomware that affected 1.5 million patient records with data like addresses, birth dates, and clinical info. Also in 2021, CaptureRx was hacked by cybercriminals who accessed 1.6 million patient records containing first name, last name, date of birth, and prescription information. These incidents reveal the need for robust security measures to safeguard patient data. Existing Regulations In response to the risks posed to patient privacy, numerous regulations have been established to set standards for protecting patient health information. In the United States, the Health Insurance Portability and Accountability Act (HIPAA) is a key legislation. HIPAA sets forth two primary rules - the Privacy Rule and the Security Rule. The Privacy Rule outlines how personally identifiable health information should be used and disclosed, whereas the Security Rule specifies the administrative, physical, and technical safeguards necessary to ensure the confidentiality, integrity, and security of electronically protected health information. To comply with HIPAA, companies must ensure their software solutions and procedures are aligned with these rules, incorporating features like strong user authentication, encryption, audit trails, automatic logoff, and secure data transmission. Regular risk assessments and staff training on HIPAA rules also form an integral part of a HIPAA compliance strategy. Across the Atlantic, the General Data Protection Regulation (GDPR) harmonizes rules for personal data protection in the European Union (E.U.), giving individuals control over their information. Other nations have enacted similar laws, including Canada’s Personal Information Protection Act (PIPA), Singapore’s Personal Data Protection Act (PDPA), and India’s Information Technology Act. While these regulations represent significant progress, there are still gaps when it comes to addressing big data: Jurisdictional conflicts can arise from countries having different standards. Stricter E.U. policies may inhibit data sharing with the U.S. Some nations lack adequate protections. Innovative technologies like AI and blockchain may outpace regulations not designed for them. Ethical dilemmas around informed consent, privacy, and fair data use require guidance. These challenges highlight the need for more adaptive laws and standards that can keep pace with the complex and rapidly evolving big data landscape. Best Practices for Data Security Data security is crucial for protecting sensitive patient information. Healthcare organizations should implement robust security controls following industry best practices: Encrypt data in transit and at rest using strong standards like AES-256 to prevent unauthorized access. Enforce multi-factor authentication and single sign-on to authenticate users and prevent account takeovers. Conduct routine security audits, penetration testing, and vulnerability scans to identify risks proactively. Implement access controls and least privilege principles to restrict data access on a need-to-know basis. Deploy data loss prevention tools to monitor and control sensitive data flows. Back up data regularly and test restores to enable quick recovery from outages or incidents. Provide comprehensive security awareness training to employees to promote best practices. Establish an incident response plan with defined procedures to contain, eradicate, and recover from breaches. Monitor systems and networks extensively using SIEM solutions to detect threats early. Ensure third-party security with vendor risk assessments and contractual protections. Following standards like NIST CSF or ISO 27001 also provides frameworks to build comprehensive security programs. With robust controls and vigilance, healthcare entities can significantly reduce data breach risks. Emerging Privacy-Enhancing Technologies Innovative privacy-enhancing technologies (PETs) can help healthcare organizations process and share data securely: Homomorphic encryption allows computations on encrypted data without decrypting it first. This enables analytics while preserving confidentiality. Differential privacy injects controlled noise into data to mask individual identities while still extracting useful insights from populations. Federated learning trains AI models across institutions without exposing raw patient data, only sharing model updates. Synthetic data generation produces artificial patient datasets with the statistical properties of real data but without being traceable to actual people. Zero-knowledge proofs let parties validate information without revealing the underlying data. Adopting PETs facilitates responsible data use for research and innovation while protecting sensitive health information. However, PETs have limitations like computational overhead, lack of standards, and re-identification risks from auxiliary data. Thoughtful implementation is key. The Patient’s Role in Data Privacy Patients have an important role in managing their health data privacy. They should: Review privacy policies to understand how providers handle sensitive information. Carefully read consent forms before permitting health data use and sharing. Request copies of medical records to review for accuracy and completeness. Ask providers about opt-out choices for data sharing like research or marketing. Enable access restrictions and notifications in patient portals to control access. Consider using personal health record apps to consolidate health data securely. Report any suspected privacy violations or breaches to providers immediately. Educating themselves on data privacy rights and options allows patients to make informed choices about their sensitive medical information. However, complex privacy policies and a lack of transparency can disempower patients. Providers should clearly communicate data practices and respect patient preferences. A Balancing Act for the Future Balancing innovation and privacy around health data requires trade-offs and diligence: Regulations must evolve to address new data sources like genomics, wearables, and social media with clear ethics-based standards. Providers should implement robust security programs while enabling controlled data use for care improvements and research. New techniques like data minimization, anonymization, abstraction, and encryption can facilitate privacy-preserving data sharing. Transparency and consent mechanisms should give patients granular controls over health data use. Cross-sector collaboration can develop best practices and governance models for secure, ethical data use. Ongoing reassessments of policies and technologies are needed as practices and risks evolve over time. With responsible data stewardship, healthcare can harness data-driven innovation to improve care and outcomes while respecting patients’ rights. Conclusion Navigating the promises and perils of big data in healthcare requires diligence and responsibility from all involved. Patients must ensure that their most sensitive information is handled with care. To be trusted by patients, healthcare providers need to be transparent, accountable, and responsible with data. Policymakers need to develop clear, adaptive guidelines for fair information use. Researchers and innovators must consider ethics alongside progress. By embracing this shared duty of care, we can unlock big data’s immense potential to understand diseases, develop treatments, and save lives - without compromising core patient rights. The following several key principles should guide us: Progress and privacy can go hand-in-hand with responsible innovation. Modern technologies can enable breakthroughs while boosting security. Listening to patient voices and concerns is crucial when shaping data practices. Policies should empower patients with control. Cross-sector and global collaboration is needed to solve a complex, international challenge. Alignment of standards serves all parties. Regular reassessment of policies, practices, and technology is necessary as the landscape evolves. What protects data today may not tomorrow. Balance, nuance, and wisdom are required around data trade-offs. Benefits and risks must be weighed carefully and holistically. The opportunities from healthcare’s digital data revolution are too great to ignore, but they must be pursued thoughtfully. With patient wellbeing at the core, a collaborative spirit, and a vision to match ambition with ethics, we can build a brighter data-driven future for human health. The potential rewards - preventing disease, expanding access, democratizing medicine - make it essential that we get big data right. By upholding both innovation and privacy, the healthcare community can earn the trust and partnership needed to transform lives with data responsibly. References The use of Big Data Analytics in healthcare | Journal of Big Data | Full Text (springeropen.com) Digital Transformation in Healthcare: Technology Acceptance and Its Applications - PMC (nih.gov) The digital transformation of healthcare organizations | Wolters Kluwer How Digital Transformation Can Improve Hospitals’ Operational Decisions (hbr.org) Digital transformation in healthcare | Deloitte Insights Digital Transformation in Healthcare in 2023: 7 Key Trends | DAP (digitalauthority.me) Frontiers | Digital transformation of health quality services in the healthcare industry during disruption and society 5.0 era (frontiersin.org) Healthcare Cybersecurity: 8 Data Protection Best Practices | CSA (cloudsecurityalliance.org) Emerging privacy-enhancing technologies: Current regulatory and policy approaches | en | OECD What is Data Privacy in Healthcare? | Box, Inc. The Balancing Act: Juggling EHRs and New Digital Health Applications | LinkedIn Five Best Practices for Securing Health Data | Persona (withpersona.com) Using Algorithmic Privacy-Enhancing Technologies in Healthcare Analytics (healthitanalytics.com)

Navigating Big Data's Potential and Privacy in Modern Medicine

AI has been slaying it in almost every sector. Healthcare is benefitting a lot from voice based . But this doesn’t mean we can trust them blindly. LLMs and ML Google DeepMind an AI program named AlphaMissense, which predicts the harm a missense mutation can do, when a single DNA letter is altered. The tool, which leverages data from 71 million single-letter mutations affecting human proteins, uses machine learning to “understand” the language of proteins, helping it score the risk level of genetic changes. created What AI is doing in synthetic biology is nothing short of revolutionary. Using AI, scientists are to create new unseen proteins like coding software. We can expect groundbreaking changes in not only medicine, but also in sectors like agriculture, energy, and climate. now programming cells Also, recently, researchers from Edith Cowan University software that quickly scans bone density scans to detect abdominal aortic calcification (AAC), which is a predictor of cardiovascular events and other health risks. The software processed images with 80% accuracy and could revolutionize early disease detection during routine clinical practice. developed In August, a ML model to identify autism spectrum disorder with 95% accuracy. While Stanford Medicine developed an algorithm that reviews images to determine a brain tumour’s aggressiveness, its genetic makeup, and effectiveness of surgery. “It’s sort of a decision support system for the physicians,” one researcher says. international researchers trained Also, Scientists at the Osaka Metropolitan University an AI model that accurately identifies cardiac functions and valvular heart diseases using chest radiographs. The research could supplement traditional echocardiography, improve diagnostic efficiency, and be especially useful in settings lacking specialized technicians. developed Startups in are also upping their game with AI. London-based Ultromics its AI to detect a heart condition that human eyes can often not detect. In July, physician-entrepreneurs at Israel’s started training AI to diagnose cardiac disease on a simple tablet, where the doctors don’t have to be heart experts. With a handheld scanner and tablet, non-cardiologists get a full diagnostic on 8 key heart readings. With its training data, the AI platform already has more experience than any living cardiologist, the physicians say. HealthTech trained Sheba Medical Center In September, , an Indian AI-powered grocery scoring and recommendation platform, tied up with, (Microbiome Research Pvt Ltd), a research platform in the field of gut microbiome and, , a diagnostics startup. With a shared commitment to preventive healthcare, the trio seeks to empower individuals with individualised in-depth diagnostics and convert that information into actionable real-time intelligence. Xume Microbiome.in TruDiagno This can help individuals to make informed choices based on the state of one’s gut microbiome and key diagnostic markers. Microbiome.in is the only private lab with an in-house Illumina and Nanopore (portable) Sequencer enabling unparalleled research on the role of gut microbiome, while TruDiagno adds advanced molecular diagnostic capabilities to the mix. Imagine the power of understanding our microbiome and diagnostic markers to drive actionable intelligence; where the answer is not giving up our favourite foods but simply making better choices based on what our gut and body are telling us Akshaye Jalan, Founder & CEO, Xume “We’re all unique, so how can a generic solution mired in restriction be the answer to good health? Imagine the power of understanding our microbiome and diagnostic markers to drive actionable intelligence; where the answer is not giving up our favourite foods but simply making better choices based on what our gut and body are telling us. This is the future of health and wellness and what our partnership with Microbiome.in and TruDiagno endeavours to do.” said Akshaye Jalan, Founder & CEO, Xume. In September, Oracle introduced voice-activated AI to hospitals, which will help doctors and clinicians to automate medical note-taking, order medications and view lab reviews while also more easily reviewing a patient’s medical record. This is not about going from 10 clicks to seven clicks or six clicks to five clicks. This is about completely eliminating clicks, where we can and where it makes sense. That is not a tomorrow or that’s years away. That is a tomorrow that’s here Mike Sicilia, executive vice president, industries at Oracle “We’ve completely rethought how long it takes somebody to get their job done using the system. I think that we are on the verge of completely eliminating ‘pajama time’ as a result. This is not about going from 10 clicks to seven clicks or six clicks to five clicks. This is about completely eliminating clicks, where we can and where it makes sense. That is not a tomorrow or that’s years away. That is a tomorrow that’s here,” Mike Sicilia, executive vice president, industries at Oracle told the audience at the . Oracle Health Conference But does this mean AI is your next doctor instead of a human? All the above instances are brilliant examples of AI helping humankind become better at healthcare, but they’re also being done under supervision of experts, who are human. The general online population has been turning to Dr. Google even before the Red Hot Chilli Peppers wrote . But there’s a new doctor in town, Dr. ChatGPT. , “the most recent version of ChatGPT made zero “grossly inaccurate” statements when presented with a standard set of eye complaints.” Californication According to NPR However, questions still remain about how to fit in LLMs into healthcare with similar safeguards that are applied when introducing new drugs or medical devices. This article was originally published by in Navanwita Bora Sachdev The Tech Panda.

Dr. ChatGPT vs. Dr. Google: An Assessment of AI Medical Advice

Software technology has come quite far indeed. I’ve been working in product management for over nine years now, with a bulk of that time being spent building platforms for software engineers and DevOps professionals. With all of that said, Even in my wildest dreams did I think that would be this accessible to the common man before 2025, yet here we are. ChatGPT, closely followed by Google Bard and other AI vendors, is changing the technological landscape in front of our eyes. Artificial Intelligence While a lot has been written on how AI can improve the efficiency of various processes in the private sector, there’s more that could be discussed on how Natural Language Processing, Machine Learning, and AI could be utilized to drive positive change in public safety. In this article, I want to think through the role of AI in helping 911, specifically in the Public Safety Answering Point (PSAP) realm. A PSAP is a call center where emergency calls are received and processed. When someone dials an emergency number, such as 911 in the United States, their call is routed to the nearest PSAP. About 911 As of 2023, there are approximately 240 million 911 calls received each year in the United States alone. This does not include the millions of non-emergency calls received each year. These calls are processed by what is called a Public Safety Answering Point, or PSAP.  The PSAP is where 911 call operators sit in, answer calls, emergency or non-emergency alike, and dispatch responses across law enforcement, fire, and EMS personnel. Now let’s talk about some of the challenges faced by 911. One of the challenges faced by 911 call centers, or PSAPs, is staffing. With millions of emergency and non-emergency calls received each year in the United States alone, it can be difficult to ensure that there are enough trained operators available to answer every call in a timely manner. This problem can be exacerbated by factors such as high turnover rates and low wages for call center staff. According to a report by the National Emergency Number Association, in 2019, approximately 25% of 911 call centers had a staffing shortage. Furthermore, the report found that the turnover rate for 911 call center staff was around 20%. These staffing challenges can result in longer wait times for callers and a higher risk of burnout for call center staff. Staffing: I : Over the years, there has been a noticeable increase in the number of emergency and non-emergency calls. According to recent statistics, the total volume of emergency calls has been rising steadily. In some regions, there has been an annual increase of approximately 3-5% in emergency calls. This upward trend indicates a growing dependence on emergency services and highlights the need for a robust system to handle emergencies effectively. Similarly, the number of non-emergency calls has also seen a significant surge. These calls typically pertain to community services, information queries, or non-urgent assistance. The exact increase varies by region, but on average, non-emergency calls have been growing by approximately 7-10%. ncreasing calls : have shown that listening to tragedy can have a lasting impact. Dr. Michelle Lilly, a clinical psychology professor at Northern Illinois University, found that between 18% to 24% of 911 dispatchers show symptoms of post-traumatic stress disorder (PTSD). That’s on par with the rate of PTSD for police officers. Mental health toll Studies We’re going to explore just some of the interesting ways PSAPs are addressing some of these challenges with Natural Language Processing, , and AI. Machine Learning AI for call routing Handling non-emergencies Non-emergency calls to 911 refer to situations in which immediate assistance from law enforcement, fire, or medical personnel is not required. These calls are made to request information, report minor incidents, or seek assistance with non-life-threatening issues. Examples of non-emergency situations that people might call 911 for include noise complaints, minor traffic accidents without injuries, reports of non-dangerous disturbances, vandalism, or lost property. While these situations are not urgent or life-threatening, they still require attention from appropriate authorities, albeit not at the same level of urgency as emergencies like crimes in progress, medical emergencies, fires, or situations with an immediate threat to safety. Estimates say that about 40% of calls to 911 are non-emergencies. The challenge is in a severely understaffed PSAP, non-emergencies could come at the cost of handling more critical calls. Non-emergency calls can sometimes flood the call center during peak hours, especially in cases where there's a community event, local news coverage, or public interest. This increased call volume can strain the resources available for addressing emergency calls promptly. Call-takers need to switch their focus between non-emergency and emergency calls, potentially causing them to lose concentration or context when handling high-stress 911 calls. This can result in errors or inefficiencies when handling emergencies. Technology can help here. One example of how AI could contribute to a solution is a virtual assistant. Imagine if every non-emergency call is picked up by an , and with a few investigative questions for triaging, could both route the call to the right representative (or guide the caller to an online reporting portal in case a human is not required to assist them). With its ability to analyze and categorize calls, the AI system could accurately direct callers to the appropriate departments or resources, ensuring a quicker response to inquiries related to non-urgent matters. AI assistant This is exactly what is doing by debuting an automated AI system. By automating the initial call-handling process for non-emergency calls, law enforcement agencies in Portland hope to optimize resource allocation and response times. The AI's capacity to swiftly process information and accurately identify the nature of the calls will contribute to more efficient utilization of personnel and departmental resources.  This innovative move has generated both curiosity and optimism within the community. While concerns regarding potential challenges in understanding nuanced human speech patterns persist, authorities have emphasized that the AI system has undergone rigorous testing to ensure its effectiveness in comprehending a wide array of caller inputs. Another example of a county doing something similar is Portland Charleston County in South Carolina. Routing medical calls Handling 911 medical calls presents a complex set of challenges for emergency response systems. The urgency of these calls demands swift and accurate assessment of medical conditions, which can be complicated by the caller's emotional state, limited medical knowledge, or the unpredictability of the situation. Call-takers must navigate through varying levels of urgency, from life-threatening emergencies to less critical situations, all while providing clear instructions and reassurance to often distressed callers. The Support Team Assisted Response (STAR) Program in Colorado and other states in the US deploys Emergency Response Teams that include Emergency Medical Technicians and Behavioral Health Clinicians to engage individuals experiencing distress related to mental health issues, poverty, homelessness, and substance misuse. STAR responds to low-risk calls where individuals are not at imminent risk. There are many places that offer star programs in Denver, Colorado, including addiction treatment centers, mental health services, and non-profit organizations. 911 could use AI similar to how Denver, CO, is utilizing the technology to analyze calls for keywords and patterns and then flag them for automatic routing to the STAR program. That could be used to help a dispatcher's ability to better identify an appropriate response since it's up to emergency dispatchers to determine whether to send police, fire, EMS, or STAR to calls. It could also help supervisors and administrators determine which calls could be STAR-eligible that aren’t being tagged as one to improve PSAP policy. AI for language translation If someone calls 911 and doesn't speak English, they will be transferred to a language line service where an interpreter will translate all questions and answers. The caller will need to tell the dispatcher what language they speak, and the call taker will conference in an interpreter to translate the conversation. The interpreter will be an over-the-phone (OPI) interpreter who is specially trained in dealing with emergency calls. Addressing non-English calls within the context of 911 services poses unique challenges to effective emergency response. Language barriers can impede clear communication between callers and call-takers, potentially leading to misunderstandings about the nature of the emergency, the location, and the required assistance. This can result in delays in dispatching appropriate resources and potentially compromising the safety of individuals in need. Moreover, the absence of readily available interpreters or translation services can further exacerbate the problem. Ensuring accurate and timely assistance for non-English callers requires specialized language support, culturally sensitive communication strategies, and the integration of technology to bridge language gaps and enhance the overall effectiveness of emergency response systems. utilizes advanced machine learning algorithms to automatically convert text or speech from one language into another, enabling effective cross-lingual communication. These systems leverage large datasets and neural networks to understand linguistic nuances, idiomatic expressions, and context, resulting in more accurate and contextually relevant translations. They continually improve over time by learning from vast amounts of multilingual content available on the internet. This technology could be leveraged in 911, and , are leading the charge on this front. AI-powered language translation cities like Baltimore, MD AI to assist in note-taking When hiring new 911 dispatchers and call-takers, a big emphasis is placed on the potential recruit’s ability to type both quickly and accurately. This is so they can distill all key insights said by a caller during an emergency as quickly as possible for first responders to take action on. In a world where 911 is facing a staffing issue, expanding the talent pool could be a key way to attract more heads into the world of public safety. AI can significantly reduce the emphasis on dispatchers taking extensive notes during a 911 call by automating the process of extracting critical information from the conversation through: AI-powered ASR technology can transcribe spoken words into text in real time. By accurately converting the caller's words into text, dispatchers can focus more on actively engaging with the caller and providing necessary guidance rather than struggling to jot down every detail. Automated Speech Recognition (ASR): Advanced AI systems can provide dispatchers with contextual insights by analyzing the call history of the caller or location. This enables dispatchers to make informed decisions based on past incidents, enhancing the quality of their responses. Companies like Prepared are pioneering technology in this field for public safety. Contextual Insights: AI can prompt dispatchers with relevant questions based on the information gathered, ensuring that no critical details are missed during the call. This guidance streamlines the conversation and reduces the need for extensive note-taking. Assistance in Information Gathering: By aiding in the process of capturing, transcribing, and extracting key information from 911 calls, AI technology allows dispatchers to focus on the human aspect of the interaction, providing empathy, reassurance, and vital instructions to callers in need while simultaneously allowing PSAPs to expand their talent pool of candidates. AI for tagging mental-health-related calls is a huge challenge in public safety, with public safety officials being exposed to some of the worst in humanity. Mental health A 911 dispatcher has to take calls back to back. In large centers, particularly where their call volume is in the hundreds of thousands in a given year, a 911 telecommunicator could spend 30 minutes on a call talking to a suicidal person and follow that up immediately by answering another call without breaks. , which is responsible for supporting more than 40 emergency communications centers across 14 counties in the Dallas-Fort Worth area of Texas, are using services from AWS to pioneer such a system. The implementation of AI software that flags distressing calls is seen as a game changer in retaining employees and ensuring their well-being. If a telecommunicator has handled a certain amount of mentally taxing calls based on keywords and sentiments analyzed, the supervisors are notified to help provide backup and support to the telecommunicator. The program will monitor calls for keywords and the tone of the caller and keep tabs on whether a call was especially stressful. Agencies like NCT911 Closing Notes In conclusion, the evolving landscape of AI technology is transforming the field of emergency response and public safety in remarkable ways. From addressing mental health challenges among 911 operators to handling non-emergency calls more efficiently, AI's potential to enhance emergency services is substantial. The intersection of AI and emergency response holds the promise of safer, more efficient, and compassionate public safety services, ushering in a new era of technological advancements that benefit communities on a global scale.

Creative Ways AI is Helping Aid Public Safety Operations

This story was originally published on by Jennifer Gollan. ProPublica Online pharmacies that sell abortion pills are sharing sensitive data with Google and other third parties, which may allow law enforcement to prosecute those who use the medications to , a ProPublica analysis has found. end their pregnancies Using , a nonprofit tech-journalism newsroom, ProPublica ran checks on 11 online pharmacies that sell abortion medication to reveal the web tracking technology they use. a tool created by the Markup Late last year and in early January, ProPublica found web trackers on the sites of at least nine online pharmacies that provide pills by mail: , , , , , , , and . Abortion Ease BestAbortionPill.com PrivacyPillRX PillsOnlineRX Secure Abortion Pills AbortionRx Generic Abortion Pills Abortion Privacy Online Abortion Pill Rx These third-party trackers, including a Google Analytics tool and advertising technologies, collect a host of details about users and feed them to tech behemoth Google, its parent company, Alphabet, and other third parties, such as the online chat provider LiveChat. Those details include the web addresses the users visited, what they clicked on, the search terms they used to find a website, the previous site they visited, their general location and information about the devices they used, such as whether they were on a computer or phone. This information helps websites function and helps tech companies personalize ads. But the nine sites are also sending data to Google that can potentially identify users, ProPublica’s analysis found, including a random number that is unique to a user’s browser, which can then be linked to other collected data. “Why in the world would you do that as a pharmacy website?” said Serge Egelman, research director of the Usable Security and Privacy Group at the International Computer Science Institute at the University of California, Berkeley. “Ultimately, it’s a pretty dumb thing to do.” Representatives for the nine sites did not respond to requests for comment. All were recommended on the popular website , which provides information about how to get abortion pills by mail, including in states where abortion is illegal. Plan C Plan C acknowledged that it does not have control over these sites or their privacy practices. While many people may assume their health information is legally protected, U.S. privacy law does little to constrain the kind or amount of data that companies such as Google and Facebook can collect from individuals. Tech companies are generally not bound by the Health Insurance Portability and Accountability Act, known as HIPAA, which limits when certain health care providers and health plans can share a patient’s medical information. Nor does federal law set many limits on how companies can use this data. Law enforcement can obtain people’s data from tech companies such as Google, whose privacy policies say the companies reserve the right to share users’ data with law enforcement. Google , which law enforcement can obtain with probable cause to believe a search is justified. requires a court order or search warrant The company received in the U.S. in 2021, the most recent year available; it does not provide a breakdown of these requests by type, such as how many involved abortion medication. more than 87,000 subpoenas and search warrants In a statement, Steve Ganem, product director of Google Analytics, said: “Any data in Google Analytics is obfuscated and aggregated in a way that prevents it from being used to identify an individual and our policies prohibit customers from sending us data that could be used to identify a user. Google has strict policies against advertising to people based on sensitive information.” Google pledged last year that it would related to people’s visits to abortion and fertility clinics, but the company has not announced any changes since then related to data involving abortion pill providers or how it handles government requests for data. delete location history data A Google spokesperson did not respond when asked whether the company has turned over any data to law enforcement about users of online pharmacies that provide abortion medication or whether it has been asked to do so. “This is problematic and dangerous — both the potential access that law enforcement has to figure out who is violating our new state bans and that we’ve let tech companies know so much about our private lives,” said Anya Prince, a law professor at the University of Iowa who focuses on health privacy. “It shows us how powerful this data is in scary ways.” Medication Abortion Using medications to induce an abortion involves taking two drugs. Mifepristone blocks the hormone progesterone, effectively stopping the growth of the pregnancy. Misoprostol, taken a day or two later, helps the uterus contract, emptying it of pregnancy tissue. This drug combination is the most commonly used method of abortion, accounting for in the U.S. more than half of abortions Demand for the drugs is expected to grow amid reproductive health clinic closures and the enactment of a cascade of state laws banning abortion since the Supreme Court overturned Roe v. Wade last June. At least 13 states now ban all methods of abortion, including medication abortion, though some allow exceptions for medical emergencies, rape or incest. People who are unable to shoulder the cost of traveling to states where abortion is legal are increasingly turning to online pharmacies to buy abortion pills without prescriptions. The mail-order pills can be taken at home, and they’re generally cheaper than abortion services provided in clinics — about $200 to $470 from online pharmacies, compared to about $500 for a first-trimester abortion conducted in a clinic. Approved by the U.S. Food and Drug Administration in 2000, mifepristone — the first tablet in the two-step regimen — can be used to help end pregnancies in their first 11 weeks. The agency initially restricted the drug, requiring patients to get it from clinicians in person. Mifepristone became more accessible during the COVID-19 pandemic, when the FDA temporarily relaxed the requirement that people visit providers in person to get the drug. The agency in December 2021, allowing people to obtain abortion medication through the mail after a telemedicine appointment. scrapped the requirement altogether Then, on Jan. 3, the FDA allowing retail pharmacies to dispense mifepristone to people who have prescriptions, potentially expanding access to medication abortion. published new rules But those rules do not help pregnant people in more than a dozen states where abortion bans prevent pharmacies from offering the drug. A week later, Alabama's attorney general said that anyone using abortion pills could be prosecuted under a state law that penalizes people for taking drugs while pregnant — despite the state’s abortion ban, which excludes abortion seekers and penalizes providers instead. He then appeared to back off his statement, saying the law would be used only to target providers. Nineteen states already , meaning people in those states must see a clinician in person or find abortion medication online on their own. Many appear ready to do the latter. ban the prescription of abortion drugs through telehealth After a draft of the Supreme Court’s abortion decision leaked last May, internet . Dozens of people have posted descriptions online of their experiences getting abortion pills, some in restrictive states. search traffic for medication abortion surged One Reddit user on an abortion subgroup in October: “I’m in TX so i ordered through abortion RX. It said it’ll be here soon like 5-6 days. I’m extremely nervous I’m doing this by myself, but I’ve looked and don’t have a lot of time to make a decision. This is the fastest way.” recounted their ordeal A New Legal Era Just two states — Nevada and South Carolina — explicitly outlaw self-managed abortion. But that hasn’t stopped prosecutors in other states from charging people for taking abortion drugs. Prosecutors have as evidence in cases charging people with illegal abortions in several states, including Georgia, Idaho and Indiana. cited online orders of abortion pills And in at least 61 cases from 2000 through 2020 spread across more than half the states in the country, prosecutors for allegedly self-managing abortions or helping someone else to do so, according to a report by If/When/How, a reproductive justice advocacy organization. investigated people or ordered their arrest In most of these cases, people had used medication for their abortions. Those prosecutors interested in criminalizing abortion are aided by state and private surveillance. “This is an entirely new era,” said Ari Waldman, a professor of law and computer science at Northeastern University. “We’re moving to a modern surveillance state where every website we visit is tracked. We have yet to conceptualize the entire body of laws that could be used to criminalize people getting abortions.” Law enforcement can use people’s behavior when visiting websites that sell abortion pills as evidence to build cases against those suspected of having abortions. Investigations and charges in these cases overwhelmingly stem from reports to law enforcement by health care providers, trusted contacts or the discovery of fetal remains, legal experts say. Once authorities launch an investigation, they can use online searches for abortion pills as part of the evidence. “This information can tell a district attorney that you went to an abortion website and you bought something,” Waldman said. “That might be enough to get a judge to get a warrant to take someone’s computer to search for any evidence related to whatever abortion-related crime they’re being charged with.” This was true even under the more limited abortion restrictions under Roe. For example, in 2017, prosecutors in Mississippi charged Latice Fisher with second-degree murder after she lost her pregnancy at 36 weeks. Prosecutors used her online search history — including a search for how to buy abortion pills online — as evidence. Fisher’s murder charge was eventually dismissed. “We have a private surveillance apparatus that is wide and is largely unregulated,” said Corynne McSherry, legal director at the Electronic Frontier Foundation, a nonprofit that promotes digital rights. “Now Google knows what you’re searching. This is a real threat. If any third party has your information, it means your data is no longer in your control and it could be sought by law enforcement. This is 100% a worry.” Opting Out Many people aren’t aware of how to opt out of sharing their data. Part of the problem is that when users visit online pharmacies that share users’ information with third parties such as Google, their information can then be shared with law enforcement if allowed by the privacy policies of those third parties. “The mere fact that you’ve used the online pharmacy to buy abortion medication, that info is now collected by Google and it is now subject to the privacy policy of Google such that you have no way of opting out of that, because it’s entirely separate from the website you went to,” Waldman said. Users can install a web browser, such as Brave or Firefox, that offers privacy protections. They can also install browser extensions to block third-party trackers and adjust the privacy settings on their browsers. But these steps . aren’t always foolproof Tech companies can still subvert them using hidden tools that users cannot see, and they likely retain vast troves of data that are beyond users’ control. “Individuals are not going to solve this problem; technical solutions aren’t going to solve this problem,” said Chris Kanich, associate professor of computer science at the University of Illinois at Chicago. “These trillion-dollar companies of the economy aren’t going anywhere. So we need policy solutions.” Congressional lawmakers have spent years discussing a national data privacy standard. The bill that has made the most progress is the . American Data Privacy and Protection Act Introduced last June by a bipartisan group of lawmakers who intended to strengthen consumer data protections, the bill limited companies from using any sensitive data, including precise geolocation information or browsing histories, for targeted advertising or other purposes. Companies would have been required to get consumers’ express consent before sharing sensitive data with third parties. The legislation passed out of its assigned House committee in July. Another bill, the , also introduced last summer, would limit the reproductive health data that companies are allowed to collect, keep and disclose. My Body, My Data Act But neither bill has passed. The My Body, My Data Act had few, if any, Republican supporters. Plus, legislators couldn’t reach an agreement over whether the American Data Privacy and Protection Act should supersede state privacy laws such as the , which provides data privacy protections for consumers in the state. California Consumer Privacy Act of 2018 Privacy experts say the most effective way to protect users’ data is for online pharmacies that sell abortion medication to stop collecting and sharing health-related data. Companies selling abortion pills should immediately stop sharing data with Google, said Cooper Quintin, senior staff technologist at the Electronic Frontier Foundation. “Web developers may not have thought they were putting their users at risk by using Google Analytics and other third-party trackers,” Quintin said. “But with the current political climate, all websites, but especially websites with at-risk users, need to consider that helping Google, Facebook and others build up records of user behavior could have a potentially horrific outcome. You can't keep acting like Roe is still the law of the land.” Are You in a State That Banned Abortion? Tell Us How Changes in Medical Care Impact You. Serious medical issues can arise during pregnancies. Our reporters want to understand how policy changes affect intimate medical decisions. Your examples can help. This story has been updated with additional comment from Google provided after publication. Update, Jan. 20, 2023: Photo by on Volodymyr Hryshchenko Unsplash

Online Pharmacies Are Sharing Your Sensitive Data 

90 Stories To Learn About Healthcare Tech

In June, the young family of generative AI (gen AI) unicorns suddenly got a new member. Inflection AI ripped the chart suggested by CBInsights just a month before and got onto the third position, dethroning Cohere AI. So we have to oblige, and after covering two Gen AI leaders OpenAI and Anthropic, we dive into the precocious startup, Inflection AI. Having introduced its core product, an AI assistant named Pi, in May and published a memo about Inflection-1, a Large Language Model ( ) behind Pi in June, Inflection AI took the industry by storm the same month with an astonishing $1.3 billion in funding and a $4 billion valuation – all within its first two years of existence! LLM The company's sudden surge owes much to the attraction of high-profile investors, including technology luminaries such as Bill Gates, Eric Schmidt, and Reid Hoffman (also a co-founder), along with tech giants and . These financial ties extend beyond mere investments. Microsoft is providing its cloud computing infrastructure to strengthen Inflection's capabilities, while Nvidia, in partnership with CoreWeave, is integrating 20,000 of its H100 GPUs for LLM training. Microsoft Nvidia Let's dive deeper and discover more about the founders, their vision, the development behind the model powering Pi, and other crucial details. Finding them was not that easy! Content Overview The starting point of Inflection Financial and legal situation Mission of Inflection and its founders First product Tech behind Pi (LLM development facts) How Inflection is going to make money? Founders’ attitude toward AI risks A few not widely known facts about Inflection's founders 1. The starting point of Inflection According to the official information, Inflection AI has three founders: Karén Simonyan (Chief Scientist and Co-founder) Reid Hoffman (Co-founder) Mustafa Suleyman (CEO and Co-founder) , Inflection hasn't been very transparent with independent media. Some information available that extends beyond mere press releases or PR work stems from the podcast hosted by Reid Hoffman for Greylock Partners (a VC fund where Hoffman has been a partner since 2009). In January 2022, announcing Mustafa Suleyman as a new partner in Greylock Partners, Reid Hoffman : “I’ve known Mustafa personally for more than a decade. We first met at a pub in London. What struck me most was our inspiring and provocative discussion on the power and potential of AI to help solve some of humanity’s most urgent challenges. It marked the beginning of a deep and lasting friendship between Mustafa and I.” Operating with a low profile said More than a decade ago from 2022 was about a time when Mustafa Suleyman (along with Demis Hassabis and Shane Legg) co-founded DeepMind, one of the first AI research labs. In 2014, it was acquired by Google ( ). In January 2022, Suleyman left Google, joined Greylock Partners, and now the main AI R&D is led by the new structure called Google DeepMind in March 2022, Inflection was born (incubated by Greylock). Located in Palo Alto, California. Introducing Inflection AI on the Greylock website, Suleyman called it an "AI-first consumer products company." However, at that moment he didn't clarify exactly what that meant. 2. Financial situation and legal situation In May 2022, in equity financing, though the source of the capital wasn't immediately clear as Inflection didn't respond to the request for more information. Subsequently, when Inflection announced the $1.3 billion funding round in 2023, they clarified in that the first round was raised from Greylock, Microsoft, Reid Hoffman, Bill Gates, Eric Schmidt, Mike Schroepfer, Demis Hassabis, Will.i.am, Horizons Ventures, and Dragoneer. TechCrunch that Inflection had secured $225 million reported the official press release At some point, the information on Inflection's website changed to describe the company as an "AI studio creating a personal AI for everyone." They chose the same legal structure as did Anthropic: , a new type of corporation to balance the interests of shareholders, materially affected individuals, and the public good. public benefit corporation 3. Mission of Inflection and its founders Introducing Inflection, as always, on the Greylock podcast, , "Our mission was to build a tool that is designed to understand the user, rather than the other way around. We wanted it – whether that be connecting with other human beings, making decisions, figuring out how to navigate a problem at work, or more.” Reid Hoffman said to be more than an answerer of questions but a participant in the dialogue that helps you along your life's path Throughout other interviews about Inflection and its first product Pi, they used words such as 'reliable,' 'ever-present,' 'non-judgmental,' 'kind,' and 'supportive’. To get to the point where a chatbot can be all this sounded like a mission. 4. First product In May 2022, Inflection AI shipped its first core product – Pi, a chatbot with the primary goal to serve as a 'Personal AI'. It’s said that Pi uses emotional intelligence to aim for human-like conversations. While its use case is narrower than other services like ChatGPT, Google’s Bard, or Microsoft’s Bing chatbot, it strives to mimic human conversation patterns. Pi's training incorporates data up to November 2022, and it is committed to user privacy, refraining from using identifiable information in training. WhatsApp, Instagram PMs, Facebook Messenger, iMessage, or via the website – the ways to talk to Pi. The main difference from the most commonly used LLM products is the Pi remembers communications with you. The chatbot is also made to resist starting to flirt with a user. Named for "personal intelligence", Pi functions in several roles like teacher, coach, and confidante, designed to align with individual interests and needs. 5. Tech behind Pi Not much is known about the technology behind Pi. However, a month after its launch, the company revealing its foundation on the proprietary LLM, Inflection-1. According to the website, Pi doesn't utilize any third-party language model APIs. released a memo In the memo, the company stated: "The Inflection-1 architecture, dataset, and training procedure are proprietary, and we omit their details in this memo." What we do know is that Inflection-1 demonstrated impressive performance on the Massive Multitask Language Understanding (MMLU) benchmark, surpassing other well-known LLMs such as Meta's LLaMA, OpenAI's GPT 3.5, and Google's PaLM 540B, particularly in answering trivia questions. Additional information from Inflection's partner, CoreWeave's official website, that Inflection AI utilized 3,500 NVIDIA H100 Tensor Core GPUs and a cluster comprising over 40,000 cables and 500 miles of InfiniBand fiber cables for MLPerf Training benchmark competition submission, achieving an impressive 11-minute training benchmark. confirms , that Inflection AI is working on a cutting-edge supercomputer equipped with 22,000 NVIDIA H100 GPUs that is expected to be one of the largest In the industry, right behind AMD's Frontier. However, there's some debate regarding the true computational power of this setup. While Inflection AI claims that these GPUs could provide 22 exaflops in 16-bit precision mode, suggest that the actual performance might not meet these expectations. There are rumors calculations by The Next Platform Though the model is proprietary, the references that are mentioned in the do not list any in-house research papers. official memo 6. How Inflection is going to make money? There are many unknowns at the moment. The developers are exploring different monetization strategies, including the possibility of adding subscription options. Another option is to keep it free but use ads to earn revenue. Currently, Inflection AI is a free platform available to everyone worldwide. On our request to learn more about the business model, neither Reid Hoffman nor Mustafa Suleyman replied. In another podcast with Reid Hoffman, Mustafa Suleyman stated, “ And likewise, we sort of don’t want to drive the kind of addictive dopamine hits of outrage or anger or despair or viral misinformation. These kinds of things have really incentivized people to spend longer on the platform, and sometimes that’s because they’re getting paid for it. We are very, very conscious that this is a moment that needs to change with AI, and that new business models need to emerge to address this." It’s really important to us that we figure out a monetization mechanic that doesn’t incentivize us to keep you engaged or maybe distracted on the platform for as long as possible. 7. Founders’ attitude toward AI risks Reid Hoffman intentionally chose to be positive about AI. , “So in recent months, He has done so in blog posts, television interviews, and fireside chats. He has spoken to government officials around the world. He hosts three podcasts and a YouTube channel. And in March, he published a book, “Impromptu,” co-written with the A.I. tool . The New York Times reports Mr. Hoffman has engaged in an aggressive thought-leadership regimen to extol the virtues of A.I. GPT-4 “I’m beating the positive drum very loudly, and I’m doing so deliberately,” he said. The 55-year-old sits on the boards of 11 tech companies including , which has gone all in on AI, and eight nonprofits. His venture capital firm, Greylock Partners, has backed at least 37 AI companies.” Microsoft , Reid Hoffman said that the , He thinks that since technology 'always puts tools in the hands of good and bad actors,' throughout our history, we have learned how to figure this out. In the same interview, he said, ' . In an interview to Washington Post conversations about superintelligence divert us from the real risk which is AI in the hands of bad actors. Part of the reason why I beat the drum so firmly in the positive camp is not that I'm risk unaware, and I think we need to navigate them, but oh, my gosh, every month we delay an AI tutor and doctor, for everyone who has a smartphone or has a friend who has a smartphone, just think about the human suffering that is caused by that delay, and that is super important This story was at Turing Post. Consider to receive new episodes straight to your inbox. originally published ing subscrib 8. A few not widely known facts about Inflection Founders Reid Hoffman Reid Hoffman ( ) is among a small group of interconnected tech executives leading the A.I. charge, many of whom also led the last internet boom. on Twitter He is part of the "PayPal Mafia," a group of former PayPal employees who went on to become successful entrepreneurs and investors in the tech industry. Reid Hoffman's entrepreneurial journey has been marked by resilience and determination. After his failure with SocialNet, he became obsessed with scaling and invested heavily in the social networking phenomenon, which eventually led to the co-founding of LinkedIn, the world's largest professional network with a diversified revenue model that includes subscriptions, advertising, and software licensing. He also led the company to its successful acquisition by Microsoft. One of the first investors in OpenAI, he was sitting on its board until March 2023, stepping down to avoid any conflicts of interest with Inflection AI. He still sits on the Board of directors at Microsoft. Reid Hoffman is a proponent of effective altruism (founders of OpenAI and Anthropic have strong ties with EA also), which advocates for using wealth and resources to create the best for the most people. In 2017, he : “If you have wealth, and at least a reasonable moral compass, to ask, “How do you create the most good for the most people?” I’m not saying this is the perfect system. But what’s the better system?” said to the Atlantic Reid Hoffman is a prominent philanthropist and has invested in impact investments through charitable entities. over $1.5 billion Hoffman is an avid fan of Dungeons & Dragons, and he attributes its influence to his thinking about life as a heroic quest with various players coming together to accomplish something. He believes in the power of friendship and considers it the true center of his life. Reid views friendship as a skill that can be learned and improved through asking questions and interacting with others. Mustafa Suleyman ( ) on Twitter He dropped out of Oxford University at 19 to start the Muslim Youth Helpline, a telephone counseling service for Muslims in the UK. He worked as a policy officer on human rights for Ken Livingstone, the Mayor of London, before founding Reos Partners, a ‘systemic change’ consultancy that uses methods from conflict resolution to navigate social problems. In February 2016, Suleyman launched DeepMind Health at the Royal Society of Medicine, a project that builds clinician-led technology for the NHS and other partners to improve frontline healthcare services He received the Commander of the Most Excellent Order of the British Empire (CBE) in 2019 for his influence in the UK technology sector and also accepted the Silicon Valley Visionary Award the same year Within DeepMind, he established the DeepMind Ethics & Society research unit, aiming to study the real-world implications of AI and promote ethical practices in the technology industry. One of the founding members of the Partnership on AI — an organization set up in September 2016 to ensure that AI is developed safely, ethically, and transparently — along with Facebook's AI head Yann LeCun, Microsoft Research director Eric Horvitz, and several others. Suleyman recently proposed a modern Turing test, , “to measure whether a machine has achieved ACI, you give an AI $100,000 and see if it can turn the seed investment into $1 million. To do so, the bot must research an e-commerce business opportunity, generate blueprints for a product, find a manufacturer on a site like Alibaba and then sell the item (complete with a written listing description) on Amazon or Walmart.com.” Suleyman expects AI will pass this more practical threshold sometime in the next two years. according to Bloomberg Karén Simonyan There is much less information about Karén Simonyan. Karén is not active on social media, he has no Twitter, a barely filled , and a well-developed . LinkedIn account Google Scholar page Education & Career History: 🔬 Chief Scientist, Inflection AI (2022 – Present) 🔬 Research Scientist, DeepMind (2014 – 2022) 🔬 Postdoc, University of Oxford (2013 – 2014) 🔬 Ph.D. student, University of Oxford (2009 – 2013) While doing his postdoc in 2013, in Oxford, Karén Simonyan along with Andrew Zisserman created the VGGNet image processing framework. VGGNet won the prestigious ImageNet Challenge. To commercialize it, they created a company Vision Factory AI, which was soon acquired by a not-very-known at that time startup DeepMind Technologies, where in 2014 Simonyan became a Principal Research Scientist. There, he played a key role in various AI breakthroughs, such as AlphaZero, AlphaFold, WaveNet, BigGAN, and Flamingo. While in DeepMind, Karen authored and co-authored numerous publications, including five papers in Nature and Science, accumulating an impressive 180,000 citations. Since he co-founded Inflection AI, only one paper was published with Karén Simonyan as a co-author . HiP: Hierarchical Perceiver That’s it for now. The information about Inflection AI will be constantly updated… Thank you for reading, please feel free to share with your friends and colleagues. This story was at Turing Post. Consider subscribing to receive new episodes straight to your inbox. originally published

The Inflection AI Story: How a Secretive Unicorn Reached a $4 Billion Valuation

Edge technology could be the key to maximizing the positive impact of for businesses by improving accessibility, performance and security. The enterprise metaverse has incredible potential in numerous industries. However, new technologies are needed to make that potential a reality. The edge computing metaverse uses the strengths of cloud computing to resolve a few key barriers to metaverse and XR adoption for businesses. the metaverse How Enterprise Will Shape the Metaverse The metaverse is often advertised as the next big technology in gaming, but many of its most promising applications are actually for businesses. The enterprise metaverse could transform the office experience, job training, remote work, and more. For example, the metaverse is highly effective for creating digital twins and simulated environments. Replicating real-world spaces and systems in the metaverse lets businesses experiment with new logistics and optimization strategies in a low-risk, cost-effective way. Studies show digital twins accounted for over revenue in 2022. By 2030, digital twin investments could total over $36 billion. $10 billion in enterprise metaverse Today, digital twins are confined to computer screens. However, the metaverse, VR and mixed reality could make them become immersive digital spaces. Edge computing would allow fast processing of real-time data that's updated immediately. The impact of the metaverse also extends to the employee experience. Businesses can use it to create virtual offices that blend in-person and remote work. They can unite employees worldwide in a shared, immersive virtual space. Employees can even get hands-on, personalized job training through simulations in the metaverse. Digital job training is currently confined to apps on a phone or laptop, leaving a gap between digital exercises and the real world. VR, AR and MR can dissolve that gap to provide a more realistic training experience. This technology is , particularly for dangerous jobs like construction and emergency response. The metaverse will expand digital immersion to the entire workplace and virtually any industry. already in development today These applications aren’t just limited to office work, either. Numerous businesses are developing revolutionary uses for the metaverse in many different industries. One startup uses the metaverse to by allowing clinicians to “scrub in” virtually from any location. expand access to health care Latency and the Edge Computing Metaverse Developers and businesses need solutions to the metaverse’s technical barriers for it to reach its full potential. Edge computing may be the key to resolving latency and bandwidth issues that hamper performance and bloat metaverse costs. Edge technology can also improve the user experience without magnifying computing power requirements. Latency and Processor Struggles Latency is one of the biggest barriers facing metaverse adoption. Expensive devices and complicated technology are among the main of VR, one of the main modes of connecting to the metaverse. VR, AR and MR headsets remain relatively rare and expensive today because the minimum technical requirements for XR are so high. factors limiting consumer adoption Headsets must render realistic graphics in 360 degrees while a user moves through a virtual space. The sheer amount of graphics to render in a VR metaverse space is more than most consumer PCs can handle. Edge technology reduces those technical capability requirements and makes it easier to deliver a good metaverse experience. Users are already seeing a precursor of edge-based XR with the emergence of cloud computing and cloud gaming. These technologies use to compensate for the lower computing power of individual devices. As a result, people can access programs and experiences that weren’t possible just a few years ago, such as playing a processor-intensive video game on a typical smartphone. computing resources from the cloud Evolving Computing to Power the Metaverse A new edge computing metaverse would bring cloud computing capabilities to XR technology. Moving computing closer to metaverse devices would allow rendering to occur faster without needing to happen entirely on the devices themselves. Metaverse platforms can use cloud computing to deliver the necessary computing power while using edge technology to bring rendering up to a speed that can provide a crisp, lifelike experience. This is important since the widespread adoption of the metaverse relies on lowering barriers to entry. Despite the massive potential for enterprise applications, the impact of the metaverse remains limited because the technology is expensive and complicated. Reducing the computing power requirements for metaverse devices would make them cheaper and easier to produce, a much more appealing investment for employers. Additionally, edge computing can provide businesses with the cybersecurity capabilities needed for mainstream metaverse adoption. Cybersecurity is a major concern for businesses today, with increasing investment in security and data privacy. 48% of CEOs reporting Security concerns will only grow more important when companies use enterprise metaverse tools and platforms for much of their daily operations. For example, a business using a metaverse office space should know its employees’ connections are secure and no one can enter the virtual workplace unauthorized. Edge technology allows businesses to keep sensitive data close to the devices where it is collected and used. Minimizing the amount of information being transmitted back and forth to a data center reduces the risk of bad actors intercepting it. The metaverse requires a massive amount of information to run effectively, particularly for applications that rely on real-world input, such as digital twins. The edge computing metaverse will make these applications more secure and streamlined, unlocking the full potential of enterprise use cases. The Impact of the Metaverse on Businesses The edge computing metaverse could make many exciting enterprise applications a reality. Cloud computing can solve the computing power challenges associated with XR devices but needs low latency to deliver a realistic experience. Edge computing can bridge that gap, providing the performance and security capabilities businesses need to supercharge enterprise metaverse applications.

The Role of Edge Computing in a Successful Enterprise Metaverse 

According to the American Medical Association, physician burnout rate reached almost 63% in 2021, while professional fulfillment scores plummeted to 22%. This seems like the right time for the medical sector to look for a solution. Whether you already have experience with AI or not, implementing natural language processing in healthcare can take some of the load off your employees’ shoulders and streamline your operations. MarketsandMarkets reports that the global natural language processing in healthcare and life sciences market was valued at $2.2 billion in 2022 and is forecast to reach $7.2 billion by 2027. And this trend will continue as more medical organizations turn to to build custom solutions. NLP service companies So, how is NLP used in healthcare, and how to deploy this technology? Keep reading to find out. What is natural language processing in healthcare? Natural language processing (NLP) is a that uses algorithms to extract meaning from unstructured human language presented either in spoken or written format. NLP requires knowledge of computational linguistics and other machine learning skills. branch of AI Natural language processing is gaining popularity in healthcare as it can analyze large quantities of unstructured medical data, such as doctor notes, medical records, clinical trial reports, and even service reviews that patients post on social media. Research shows that about and not utilized to its full potential. NLP can change that by analyzing data and extracting insights to guide doctors and pharmacists to make more informed decisions. 80% of healthcare data is unstructured Here are some of the main NLP techniques that are useful in healthcare settings: . OCR converts printed and handwritten text to a machine-readable format. It can extract text from images and tables and pass it to other NLP algorithms for further processing. If you want to learn more about the topic, check out our recent . Optical character recognition (OCR) guide on optical character recognition . This approach involves assigning semantic labels based on predefined categories. For instance, it allows doctors to label a patient as “at risk of hospitalization” based on certain keywords in their medical notes. Text classification . This is an information extraction model that can pinpoint entities in text. Physicians can use this technique to extract entities, such as “treatment” and “symptoms,” from a heap of medical data. Named entity recognition . This technique can sort information based on common topics. For example, it can group together all doctors’ notes on patients suffering from a particular disorder. Topic modeling . This NLP method can establish semantic relationships between entities. For instance, it can point out that a particular patient was treated at that hospital under the supervision of that doctor. Relationship extraction NLP use cases in healthcare After learning about what healthcare NLP is and how it works, let’s see how it can contribute to the healthcare field. Clinical documentation management Research shows that physicians for every patient they see. This time could have been spent caring for patients. Instead, it’s wasted on the screen and contributes to doctor burnout. spend 16 minutes on EHRs on average Enhancing clinical documentation through speech recognition Natural language processing in healthcare can help physicians work with electronic health records (EHRs). Doctors can use speech-to-text conversion tools with built-in NLP capabilities to transcribe their notes and enter them into the corresponding patent’s EHR fields. Also, medical staff can query the NLP tools to extract relevant data from EHRs. For instance, Nuance’s solution is a cloud-based speech recognition tool that helps medical personnel document patients’ stories in their EHRs. The company claims that deploying their solution will cut time spent on clinical documentation by half. Concord Hospital deployed Dragon Medical One, and more accurate clinical documentation. Dragon Medical One 75% of employees reported Supporting doctors in decision making NLP solutions can also analyze clinical documents and support physicians in real-time decision making. For instance, after analyzing doctor notes, the system can predict hospital bed demands, which gives hospital staff time to prepare and accommodate incoming patients. Medical coding and billing Medical coding means deriving billable information from clinical notes and transferring it into standardized medical codes. Traditionally, a human coder would perform this task. But manual coding is slow and prone to error, meaning that the provider might not be able to claim and receive the full amount in compensations. NLP-powered computer-assisted coding (CAC) tools can derive information from doctor notes and patient EHRs about different therapies and procedures they encountered, and provide the corresponding insurance codes to solidify claims. One example of CAC natural language processing in healthcare is . The company employs over 150 coding experts and remains up to date with the newest regulations. This solution assists coders by aggregating and analyzing patient documentation, offering auto-suggested tags and comprehensive review and approval tools. 3M 360 Encompass System Clinical trial management Studies reveal that around because they can’t recruit enough participants. Medical NLP can speed up recruitment by scanning patients’ medical data in search of eligible trial candidates. The technology can also help people who want to participate in clinical trials. It can transform eligibility criteria into questions, allowing candidates to filter through many trials fast.Headquartered in Canada, Inspirata built a solution that employs natural language processing in healthcare to . Its NLP engine mines clinical documents and transferred unstructured data into structured insights on patient eligibility. 80% of clinical trials are delayed or terminated facilitate patient recruitment Another successful NLP platform, , can not only help recruit patients but also assist in trial design and site selection. It can parse a variety of sources, such as news feeds, patents, medical literature, and similar trial descriptions. Linguamatics NLP Refer to our to discover how other subdisciplines of AI contribute to this field. article on AI in clinical trials Patient sentiment analysis Applying NLP in healthcare helps medical staff aggregate and analyze customer reviews and opinions from different social media platforms. Natural language processing tools can process thousands of reviews to gauge patients’ attitude towards the care they received. Such tools can identify aspects that frustrate patients, segment them based on frequency, and start implementing improvements for the most recurring negative feedback. For instance, a comment such as “ ” indicates a negative experience with the administrative staff. I waited for 30 minutes on the phone to talk to the receptionist Another way of benefiting from healthcare NLP in sentiment analysis is processing survey results. Instead of relying on social media, healthcare facilities can compose targeted surveys, distribute them among patients, and use NLP solutions to automatically review every response. In a recent study, a team of researchers that can scan survey results to determine how patients feel about their healthcare provider. The team designed a survey with open-ended questions, distributed it on vestibular schwannoma patients, and received 534 responses. The algorithm classified the responses into two groups — positive and negative sentiments. This classification had a 90% overlap with the manual classification produced by human experts. built an NLP solution Check out our blog article if you want to learn more about . how sentiment analysis can boost your customer service Drug discovery To find effective candidate drugs, scientists need to identify the biological origins and understand the disease first. This includes reviewing large amounts of medical literature, patient data, and more. Natural language processing in healthcare can quickly sift through all this information, extract what is relevant, and present it to researchers in a compelling format so that they can learn about similar diseases and how they were treated. For instance, Texas-based offers an AI-powered software that can assist in drug discovery by applying NLP to analyze unstructured data from multiple sources, such as gene cards, PubMed publications, proprietary global research data repositories, clinical trial documentation, and more. After absorbing the available disease-related knowledge, the tool builds an ontology of existing drugs that highlights promising research areas to discover new effective drugs. Lymba Additionally, NLP can analyze social media feeds and patients’ medical records to gauge adverse effects of different drugs. Some tools can even correlate a drug’s side effects to the dosage consumed and the frequency of use. For more exciting applications of you can refer to our blog. AI in drug discovery Challenges of NLP implementation in healthcare Applying natural language processing in healthcare has many benefits. But if you decide to enhance your practice with this technology, here are the challenges that you are likely to encounter. Specific language requirements of the healthcare field Doctors have their specific way of writing clinical notes and other medical documents. This style is typically heavy with jargon, acronyms, and abbreviations, and has limited context. To add to the confusion, a third of the clinical abbreviations present in the Unified Medical Language System Metathesaurus . have several meanings each Even words that are spelled out in full can be confusing. For instance, “discharge” can either mean release from a hospital or bodily fluids, depending on the context. And if you want to extend the application of NLP in healthcare beyond your clinic and use it to analyze patient’s feedback on social media, then you need to familiarize it with the specific abbreviations and emoticons that people use to express their emotions instead of words. If you acquire ready-made general-purpose NLP algorithms, you will need to retrain them to operate in healthcare settings. And even better, familiarize them with the vocabulary used by your doctors. Issues related to human language complexity It’s relatively easy to deploy NLP algorithms to spot terms that are explicitly mentioned and unambiguous. But human language is not always that straightforward. Sometimes, algorithms will need to deal with the following: . Natural language processing models in healthcare have to be able to infer information from the input text. For example, if you want to know whether a patient has social support, there might not be a clear mention of this in their medical notes. However, there might be something like “relative at bedside,” that implies the patient has a form of support. Inferences . A basic keyword-driven NLP will perceive the two statements “wife helps patient with medication” and “patient helps wife with medication” as identical, while in reality the semantics in both cases are very different. Semantics beyond keywords . Clinicians often use negation to indicate absence of medical conditions. For example, a doctor might write “a brain malignancy was ruled out” to document the fact that a malignant brain tumor wasn’t found. Medical NLP algorithms can’t mistake this with the presence of the disease. Negation Decide on how advanced your algorithms need to be based on the target business cases. And if you opt for more basic solutions, be aware of their limitations. Bias and explainability Like any other AI technology, NLP in the medical field can acquire during training on unbalanced datasets or when it continues to learn on the job. Gender bias is one of the most prominent types in natural language processing-based tools. For example, with jobs that require high-level education, such as a doctor, while females are connected to less knowledge-intensive occupations, like a nurse. different types of bias GPT-3 tends to associate males Unfortunately, AI bias is not uncommon in healthcare. For example, an algorithm that was supposed to determine the severity of patients’ illness required for black patients just because, historically, more money was spent on white patients’ needs. significantly downplayed the level of care It can be hard to spot bias when the algorithms are “black box” models that don’t explain how they arrive at their conclusions. One solution to this issue is to deploy . With this technology, healthcare natural language processing tools justify their recommendations, allowing you to verify whether they are biased or not. explainable AI Another way to spot bias is using a curated dataset that scientists designed to detect specific types of bias. This solution is not scalable to large applications but works well with limited usage. In addition to the three NLP-specific challenges presented above, you might encounter general obstacles associated with implementing any AI-powered technology, such as: . Many healthcare organizations still rely on outdated legacy systems that aren’t built to work with AI-driven technology and deal with large amounts of data. Integration with legacy systems . Medical facilities struggle to provide sufficient, high-quality data that faithfully represents the target population. Insufficient training data . Hospitals struggle with the responsibility that comes with using AI, such as who has the final say on selecting the right treatment, and who is responsible if the treatment didn’t work. Ethical considerations and moral hazards For a more detailed explanation, check our article on . top AI implementation challenges Step-by-step implementation of natural language processing in healthcare If you want to use NLP in healthcare, you can follow these steps: . Determine which problems you want to solve with NLP and whether you need advanced language processing capabilities, such as negation and inference. Identify potential use cases . Decide if you want to acquire a ready-made NLP tool or build your own customized product. Off-the-shelf solutions are suited for companies with limited funds and flexible internal processes that can be adapted to accommodate an external product. Consider a custom-built solution if you have rigid workflows, are looking for something scalable, and need to integrate the new tool with legacy systems. In the latter case, you will need to look for a trusted to collaborate with. Build/buy an NLP solution AI healthcare solutions company . Even if you opted for a ready-made solution in the previous step, you will still need to retrain it for optimal performance, as mainstream NLP models are not designed to handle the specifics of the healthcare sector. You will need to that is bias-free and representative of your target population. Prepare the training dataset prepare a training dataset . Test the tool for bias, usability, and how it integrates with the rest of the workflow. Train and validate the model . If you want to build and/or adopt AI models in the medical sector, compliance is of utmost importance. Your tools need to comply with the US Food and Drugs Administration (FDA), the General Data Protection Regulations (GDPR), and any other regulatory bodies in the countries of your operations. You can find more about the topic in our recent article on . Make sure your algorithms are compliant healthcare IT standards . If you and hired a tech vendor, they will design and build the NLP tool with the specifics of your system in mind. They will also help you integrate the solution with your legacy systems. Integrate with your existing systems opted for a custom healthcare solution . The effort doesn’t end at deploying natural language processing in healthcare. You need to audit the NLP models to make sure they are still producing the desired results and update them to reflect any changes in the target use cases. Monitor and adjust the algorithms constantly Concluding thoughts NLP has many applications in healthcare. It can help save time on medical coding, facilitate recruiting participants for clinical trials, keep patients happy through sentiment analysis, and much more. To successfully deploy natural language processing in healthcare, dedicate time to finding a trusted who has experience in the medical field. A tech partner will help you with: machine learning vendor Creating and training custom NLP models that: Are free from bias Fit seamlessly with the rest of your system Tailor to your practice’s vocabulary and have the desired degree of complexity Complying with the privacy requirements of the healthcare sector Supporting you in auditing and updating the algorithms when needed Looking to enhance your medical practice with NLP? ! We will help you build a compliant solution tailored to your internal systems and vocabulary. Drop us a line Also published . here

Natural Language Processing in Healthcare: A Path to Adoption

Procesamiento del lenguaje natural en el cuidado de la salud: un camino hacia la adopción

Traitement du langage naturel dans les soins de santé : une voie vers l'adoption

हेल्थकेयर में नेचुरल लैंग्वेज प्रोसेसिंग: ए पाथ टू एडॉप्शन

医療における自然言語処理: 導入への道

Processamento de linguagem natural na área da saúde: um caminho para a adoção

Xử lý ngôn ngữ tự nhiên trong chăm sóc sức khỏe: Con đường dẫn đến việc áp dụng

医疗保健中的自然语言处理：采用的途径

Take a look at these statistics: Artificial intelligence can according to a recent study by Insider Intelligence reduce drug discovery costs by up to 70% AI can on clinical research while cutting the phase’s length by half save pharma $28 billion a year AI can spare pharma companies on R&D costs up to $54 billion annually It’s not surprising that pharmaceutical companies to cut the costs and time required for drug development. The global AI in pharma market and is estimated to surpass $9,241 million by 2030, growing at a CAGR of 29.4%. turn to medical AI services was valued at around $905 million in 2021 Motivated to learn more about how using artificial intelligence in the pharmaceutical industry can improve your drug development processes? Then continue reading. 5 top use cases of AI in the pharma sector You will notice that it’s common for pharma companies to team up with tech innovators to successfully deploy AI. Accenture conducted a survey where in profit after partnering with a tech vendor, with 76% of pharma executives citing effective partnership as a key success factor. 61% of the respondents reported at least a 5% increase Here are 5 top applications of artificial intelligence in pharmaceutics. Drug discovery The Congressional Budget Office reports that the R&D costs of , which includes research and clinical trials. developing a new drug can exceed $2 billion Deploying AI in pharma enables researchers to sift through enormous datasets, such as small molecule libraries and spot disease patterns, and learn which chemical compositions can be a good fit for various biological targets. AI can generate chemical compounds either as a text string or as a graph architecture. It’s important to validate the resulting compounds, as many of them will not make sense, could be toxic, or could contain a component that shouldn’t be a part of any drug. In addition to discovering candidate compositions, scientists can use AI algorithms to parse medical literature on how to best synthesize the drug and design clinical trials. Research shows that pharmaceutical artificial intelligence can , saving the pharma sector up to $26 billion in annual expenses. cut drug synthesizing and screening time by 50% There are many great examples of pharma companies to facilitate drug discovery. For instance, GSK, a British pharmaceutical company headquartered in London, partnered with California’s Vir Biotechnology during the pandemic with the help of AI and a human gene editing tool, CRISPR. Vir already had an antibody platform that it deployed to discover drugs for different respiratory pathogens in the past. And now, in this collaboration, they discovered sotrovimab, an antibody to neutralize COVID-19. deploying AI solutions to accelerate COVID-19 antibody discovery that binds to a SARS-CoV-2 epitope In another example of collaboration between Europe and the US, a French pharma and healthcare company Sanofi partnered with California-based biotech innovator Atomwise to discover and synthesize drug compounds for five different targets. Sanofi wanted to steer clear of the traditional drug discovery approach and for their innovation and AI capabilities. paid Atomwise $20 million upfront Clinical trials AI has many . One of them is identifying the right candidate participants. The technology can analyze patient data, genetic information, doctor notes, and other information, and pick people who are eligible for a particular trial. AI can even help decide on the optimal population size based on the existing description of similar trials. applications in clinical trials within their target time frame. due to recruitment-associated challenges. 86% of clinical trials fail to recruit enough patients One-third of phase Ⅲ clinical trials have to stop For instance, relies on analytics and natural language processing (NLP) to analyze patient information. The tool can handle unstructured data, like doctor’s notes, and produce an insightful patient summary. Clinical researchers use these highlights to select and recruit patients. IBM Watson As AI helps pharma companies to find patients, it also works the other way around. , a clinical trial patient recruitment platform, uses NLP to analyze their text and screen them for trial inclusion/exclusion criteria. It requires patients to answer a few simple questions on its platform and suggests a list of trials that the person can join. Antidote Drug manufacturing Deploying AI in the pharmaceutical industry offers multiple opportunities to improve the drug production process. The technology could: . AI can inspect drugs on the conveyor belt and spot defects, such as damaged packaging. Moreover, the technology can identify any potential issues by analyzing manufacturing data, like quality control tests. For instance, AstraZeneca employs looking for defects, while Merck applies AI to spot problems in vaccine vials. Assist in drug quality control machine learning to analyze drug images . AI can monitor equipment on the production line and identify potential defects through sensors that measure equipment vibration, temperature, sound, etc. This gives employees time to fix the device before it breaks down, halting production. Facilitate predictive maintenance . AI can analyze data on energy consumption, raw material waste, and other parameters and put forward recommendations on how to . Also, the technology can predict demand, so that pharma manufacturers avoid producing large quantities of drugs that will not be consumed and will otherwise go to waste. Cut down on material waste improve the manufacturing process Drug marketing The pharma sector largely depends on sales. Companies aim to reach as many customers as possible while offering a distinctive user experience and a customized approach. Artificial intelligence in pharma can facilitate drug marketing by: Comparing past marketing campaigns and identifying the most profitable approaches. The technology can also analyze customer engagement tactics and pick the most successful ones. Aggregating customer data in real time to understand their behavior and what they are looking for to create a tailored advertisement. Optimizing pricing of new drugs by considering all involved stakeholders and data on similar medications. Simulating different market scenarios by forecasting changes in demand, competitor behavior, etc. This allows pharmaceutical firms to be prepared for sudden landscape changes. Finding new consumers for existing drugs. For instance, for Chantix (a drug that helps people quit smoking). The tool analyzed data from the Centers for Disease Control and Prevention to identify previously untapped population segments. Pfizer relied on AI to spot and reach new potential customers Drug dosage optimization AI can analyze large quantities of unstructured patient data and calculate the optimal dosage of a particular drug for this person to achieve the best possible results with minimal side effects. Artificial intelligence models in the pharma industry can analyze the following information: Medical history, such as doctor’s notes, lab test results, genetic makeup Medical images, such as Magnetic Resonance Imaging (MRI) scans Biomarkers, such as protein levels and genetic mutations Drug characteristics, such as its metabolism Potential side effects of a drug and of similar drugs When the optimal dosage is calculated, the technology can monitor its effectiveness and make adjustments when needed. To give a real-life example, a California-based company Dosis built an AI-driven personalized medicine dosing platform that dialysis clinics can use to manage chronic drug intake. In his interview with HealthcareITNews, Dosis’ CEO Shivrat Chhabra mentioned this platform helped clients while improving patient outcomes. reduce drug consumption by 25% Challenges associated with implementing AI in pharma Some of these obstacles are specific to the field, and some are more general and apply to all projects involving this technology. One of the key challenges is the . This is particularly hard as the expenses associated with drug development are already rather high. You can turn to experienced AI consultants to learn how to cut down on costs and still get a viable product. enormous costs associated with artificial intelligence Here are other prominent challenges that you can face during pharmaceutical AI implementation. Data quality and quantity According to a recent study by McKinsey, the on the way to applying analytics in the healthcare field. lack of integrated data sources was the chief obstacle Pharma AI models typically require large datasets to learn. However, it’s a challenge to obtain a sufficient dataset for each disease, especially the rare ones. So, as training datasets are getting smaller, the data that an AI-powered drug development tool has to handle is rather complex. Think of patient data. It includes historical information, genetic makeup, doctor notes, medical scans, etc. Under these conditions, it’s a challenge to build accurate algorithms. When training data is lacking, it’s possible to use synthetic data generators for some pharma applications. For instance, suitable for pharmaceutical usage. Healthcare data is among the most sensitive data types, and privacy is of the essence in such applications. Synthetic datasets can solve this issue. As Andreas Ponikiewicz, VP of Global Sales at Mostly AI, puts it, Mostly AI claims it can generate data “With generative AI based synthetic healthcare data, that contains all the statistical patterns, but is completely artificial, the data can be made available without privacy risk.” Another option for acquiring data for experimenting with AI and pharma is to become a part of a specialized collaboration. For example, the Massachusetts Institute of Technology initiated the Machine Learning for Pharmaceutical Discovery and Synthesis Consortium. 13 pharma companies joined the consortium to design and build AI algorithms for small molecule discovery. You need to make sure that the data used within pharmaceutical applications is all realistic. But it’s rather costly to verify that, as it requires the intervention of human experts. Lack of interoperability and a unified data standard There are still multiple , which means that each hospital can adopt a standard of their choice for data storage and formatting. This makes it hard to integrate and use patient data needed for drug-related research from different medical facilities. healthcare IT standards and regulations These issues of AI in the pharma industry can be addressed on the governmental level. For instance, the Swiss Personalized Health Network (SPHN) is a health data unifying initiative by the Swiss government. The that streamlines medical data exchange among Swiss hospitals, research institutes, and regulatory bodies. SPHN was set to build a national infrastructure On an individual level, pharma researchers can benefit from platforms like , which uses NLP to scan and extract data from heterogeneous electronic health record (EHRs) systems. Deep 6 AI Algorithmic bias “All data is biased. This is not paranoia. This is fact.” – Dr. Sanjiv Narayan, professor of medicine at Stanford University. if their training dataset wasn’t representative of the target population. Data bias has specifically been a problem in the pharmaceutical and healthcare sectors. Research shows that only a offer evidence on covering the bias issue. AI-powered models can easily develop bias few AI-powered products submitted for FDA approval Some medical professionals if data scientists work more closely with clinicians and learn more about data while building the algorithms. They can request information, such as where the data came from and what was the original goal of gathering it. Then engineers can make tweaks to the algorithms to address any population misrepresentation. believe that it will help reduce bias Algorithms can also acquire bias as they continue to learn on the job. Hence, systematic audits are essential to ensure that all AI-based tools are still relevant and work as expected. Integration with existing systems Deploying AI in pharma implies integrating it with the existing platforms and applications. Many pharma companies still rely on outdated legacy systems that are not designed to work with AI or deal with a large amount of data. Such systems use their proprietary protocols and are hard to . integrate with modern applications Pharma companies that want to use modern technology alongside legacy systems can benefit from designed to fit seamlessly with the existing legacy systems. custom pharma software solutions The complexity of the pharmaceutical applications The use cases of artificial intelligence in the pharmaceutical industry are rather complex, and there is a large room for error in the predictions that the technology makes. Here is what makes pharma so intricate: Every patient has individual characteristics and many factors to consider in clinical trials. If you are developing a drug for liver-related issues, you need to find trial participants with no other health conditions that can influence and sway your results. The need to consider the interaction between different drugs as one person might be taking multiple drugs to treat different conditions. Disease variability as one medical condition can have several variants and manifest itself in different ways. Training datasets describing diseases and treatments are not balanced, which can force the algorithm to recommend the most frequently occurring solution even if it’s not the correct one. To sum it up Deloitte reports that have witnessed some progress over the past years. And the consultancy believes AI in pharma can change this. In addition to the applications mentioned above, , which is vital in this field. only a few of the 7,000 rare diseases that we know AI can help pharma companies achieve compliance If you want to incorporate this advanced technology into your business, you are likely to have to team up with a tech vendor of your choice. Also, it’s a good practice to: Make sure your training dataset is realistic, even if the verification process is expensive and requires human expert intervention Incorporate AI into your strategy instead of treating it as a side project Build strong AI skills or outsource this to dedicated teams Encourage a close collaboration between your data scientists and clinicians Beware of the latest regulations regarding using AI in pharma, as these are changing rapidly Build your own ethical expertise to address any concerns associated with AI and pay a special attention to privacy and security if you decide to collaborate with other players in the field and share your data Regularly monitor the algorithms’ performance for bias and inaccuracy, be it for disease discovery, recruitment of trial participants, or even drug advertising. For example, the University of California on social media and discovered that AI models tend to excessively recommend these drugs to Latino and African-Americans customers. conducted a study on advertising mental health medication When using AI to generate chemical compounds, always validate the results, as it can deliver toxic or otherwise unsuitable components Speaking of the future of artificial intelligence in the pharmaceutical industry, that will include the following players: PwC predicts the emergence of a new digital health ecosystem Solution vendors, who will offer personalized treatments, drug dosages, etc. Orchestrators, who can use AI and analytics to address patients’ needs Platform providers, who will mediate between the aforementioned players And according to the consultancy, firms who will still refuse to make AI a part of their operations will turn into a mere “contract manufacturers” for the rest of the ecosystem. So, if you haven’t yet considered enhancing your business processes with AI, this seems like a good time to experiment with the technology. Are you looking to save time and money on drug development and clinical trial organization? ! We will help you build and train AI models and integrate them seamlessly into your system. Drop us a line

A Pill a Day: What AI Can REALLY Do in the Pharma Sector

Una pastilla al día: lo que la IA REALMENTE puede hacer en el sector farmacéutico

Une pilule par jour : ce que l'IA peut VRAIMENT faire dans le secteur pharmaceutique

ए पिल ए डे: एआई वास्तव में फार्मा सेक्टर में क्या कर सकता है

A Pill a Day: AI が製薬業界で本当にできること

Uma pílula por dia: o que a IA REALMENTE pode fazer no setor farmacêutico

Một viên thuốc mỗi ngày: AI THỰC SỰ CÓ THỂ LÀM GÌ TRONG NGÀNH DƯỢC PHẨM

一天一粒：人工智能在制药行业的真正作用

Studies in the Psychology of Sex, Volume 2 by Havelock Ellis is part of the HackerNoon Books Series. You can jump to any chapter in this book here . THE STUDY OF SEXUAL INVERSION II.—THE STUDY OF SEXUAL INVERSION. Westphal—Hössli—Casper—Ulrichs—Krafft-Ebing—Moll—Féré—Kiernan—Lydston—Raffalovich—Edward Carpenter—Hirschfeld. Westphal, an eminent professor of psychiatry at Berlin, may be said to be the first to put the study of sexual inversion on an assured scientific basis. In 1870 he published, in the Archiv für Psychiatrie, of which he was for many years editor, the detailed history of a young woman who, from her earliest years, differed from other girls: she liked to dress as a boy, only cared for boys' games, and as she grew up was sexually attracted only to women, with whom she formed a series of tender relationships, in which the friends obtained sexual gratification by mutual caresses; while she blushed and was shy in the presence of women, more especially the girl with whom she chanced to be in love, she was always absolutely indifferent in the presence of men. Westphal—a pupil, it may be noted, of Griesinger, who had already called attention to the high character sometimes shown by subjects of this perversion—combined keen scientific insight with a rare degree of personal sympathy for those who came under his care, and it was this combination of qualities which enabled him to grasp the true nature of a case such as this, which by most medical men at that time would have been hastily dismissed as a vulgar instance of vice or insanity. Westphal perceived that this abnormality was congenital, not acquired, so that it could not be termed vice; and, while he insisted on the presence of neurotic elements, his observations showed the absence of anything that could legitimately be termed insanity. He gave to this condition the name of "contrary sexual feeling" (Konträre Sexualempfindung), by which it was long usually known in Germany. The way was thus made clear for the rapid progress of our knowledge of this abnormality. New cases were published in quick succession, at first exclusively in Germany, and more especially in Westphal's Archiv, but soon in other countries also, chiefly Italy and France. While Westphal was the first to place the study of sexual inversion on a progressive footing, many persons had previously obtained glimpses into the subject. Thus, in 1791, two cases were published of men who showed a typical emotional attraction to their own sex, though it was not quite clearly made out that the inversion was congenital. In 1836, again, a Swiss writer, Heinrich Hössli, published a rather diffuse but remarkable work, entitled Eros, which contained much material of a literary character bearing on this matter. He seems to have been moved to write this book by a trial which had excited considerable attention at that time. A man of good position had suddenly murdered a youth, and was executed for the crime, which, according to Hössli, was due to homosexual love and jealousy. Hössli was not a trained scholar; he was in business at Glarus as a skillful milliner, the most successful in the town. His own temperament is supposed to have been bisexual. His book was prohibited by the local authorities and at a later period the entire remaining stock was destroyed in a fire, so that its circulation was very small. It is now, however, regarded by some as the first serious attempt to deal with the problem of homosexuality since Plato's Banquet. [114] [115] Some years later, in 1852, Casper, the chief medico-legal authority of his time in Germany,—for it is in Germany that the foundations of the study of sexual inversion have been laid,—pointed out in Casper's Vierteljahrsschrift that pederasty, in a broad sense of the word, was sometimes a kind of "moral hermaphroditism," due to a congenital psychic condition, and also that it by no means necessarily involved sodomy (immissio penis in anum). Casper brought forward a considerable amount of valuable evidence concerning these cardinal points, which he was the first to note, but he failed to realize the full significance of his observations, and they had no immediate influence, though Tardieu, in 1858, admitted a congenital element in some pederasts. [116] The man, however, who more than anyone else brought to light the phenomena of sexual inversion had not been concerned either with the medical or the criminal aspects of the matter. Karl Heinrich Ulrichs (born in 1825 near Aurich), who for many years expounded and defended homosexual love, and whose views are said to have had some influence in drawing Westphal's attention to the matter, was a Hanoverian legal official (Amtsassessor), himself sexually inverted. From 1864 onward, at first under the name of "Numa Numantius" and subsequently under his own name, Ulrichs published, in various parts of Germany, a long series of works dealing with this question, and made various attempts to obtain a revision of the legal position of the sexual invert in Germany. Although not a writer whose psychological views can carry much scientific weight, Ulrichs appears to have been a man of most brilliant ability, and his knowledge is said to have been of almost universal extent; he was not only well versed in his own special subjects of jurisprudence and theology, but in many branches of natural science, as well as in archeology; he was also regarded by many as the best Latinist of his time. In 1880 he left Germany and settled in Naples, and afterward at Aquila in the Abruzzi, whence he issued a Latin periodical. He died in 1895. John Addington Symonds, who went to Aquila in 1891, wrote: "Ulrichs is chrysostomos to the last degree, sweet, noble, a true gentleman and man of genius. He must have been at one time a man of singular personal distinction, so finely cut are his features, and so grand the lines of his skull." [117] [118] For many years Ulrichs was alone in his efforts to gain scientific recognition for congenital homosexuality. He devised (with allusion to Uranos in Plato's Symposium) the word uranian or urning, ever since frequently used for the homosexual lover, while he called the normal heterosexual lover a dioning (from Dione). He regarded uranism, or homosexual love, as a congenital abnormality by which a female soul had become united with a male body—anima muliebris in corpore virili inclusa—and his theoretical speculations have formed the starting point for many similar speculations. His writings are remarkable in various respects, although, on account of the polemical warmth with which, as one pleading pro domo, he argued his cause, they had no marked influence on scientific thought. [119] This privilege was reserved for Westphal. After he had shown the way and thrown open his journal for their publication, new cases appeared in rapid succession. In Italy, also, Ritti, Tamassia, Lombroso, and others began to study these phenomena. In 1882 Charcot and Magnan published in the Archives de Neurologie the first important study which appeared in France concerning sexual inversion and allied sexual perversions. They regarded sexual inversion as an episode (syndrome) in a more fundamental process of hereditary degeneration, and compared it with such morbid obsessions as dipsomania and kleptomania. From a somewhat more medico-legal standpoint, the study of sexual inversion in France was furthered by Brouardel, and still more by Lacassagne, whose stimulating influence at Lyons has produced fruitful results in the work of many pupils. [120] Of much more importance in the history of the theory of sexual inversion was the work of Richard von Krafft-Ebing (born at Mannheim in 1840 and died at Graz in 1902), for many years professor of psychiatry at Vienna University and one of the most distinguished alienists of his time. While active in all departments of psychiatry and author of a famous textbook, from 1877 onward he took special interest in the pathology of the sexual impulse. His Psychopathia Sexualis contained over two hundred histories, not only of sexual inversion but of all other forms of sexual perversion. For many years it was the only book on the subject and it long remained the chief storehouse of facts. It passed through many editions and was translated into many languages (there are two translations in English), enjoying an immense and not altogether enviable vogue. Krafft-Ebing's methods were open to some objection. His mind was not of a severely critical order. He poured out the new and ever-enlarged editions of his book with extraordinary rapidity, sometimes remodelling them. He introduced new subdivisions from time to time into his classification of sexual perversions, and, although this rather fine-spun classification has doubtless contributed to give precision to the subject and to advance its scientific study, it was at no time generally accepted. Krafft-Ebing's great service lay in the clinical enthusiasm with which he approached the study of sexual perversions. With the firm conviction that he was conquering a great neglected field of morbid psychology which rightly belongs to the physician, he accumulated without any false shame a vast mass of detailed histories, and his reputation induced sexually abnormal individuals in all directions to send him their autobiographies, in the desire to benefit their fellow-sufferers. It is as a clinician, rather than as a psychologist, that we must regard Krafft-Ebing. At the outset he considered inversion to be a functional sign of degeneration, a partial manifestation of a neuropathic and psychopathic state which is in most cases hereditary. This perverse sexuality appears spontaneously with the developing sexual life, without external causes, as the individual manifestation of an abnormal modification of the vita sexualis, and must then be regarded as congenital; or it develops as a result of special injurious influences working on a sexuality which had at first been normal, and must then be regarded as acquired. Careful investigation of these so-called acquired cases, however, Krafft-Ebing in the end finally believed, would indicate that the predisposition consists in a latent homosexuality, or at least bisexuality, which requires for its manifestation the operation of accidental causes. In the last edition of his work Krafft-Ebing was inclined to regard inversion as being not so much a degeneration as a variation, a simple anomaly, and acknowledged that his opinion thus approximated to that which had long been held by inverts themselves. [121] At the time of his death, Krafft-Ebing, who had begun by accepting the view, at that time prevalent among alienists, that homosexuality is a sign of degeneration, thus fully adopted and set the seal of his authority on the view, already expressed alike by some scientific investigators as well as by inverts themselves, that sexual inversion is to be regarded simply as an anomaly, whatever difference of opinion there might be as to the value of the anomaly. The way was even opened for such a view as that of Freud and most of the psychoanalysts today who regard a strain of homosexuality as normal and almost constant, with a profound significance for the psychonervous life. In 1891 Dr. Albert Moll, of Berlin, published his work, Die Konträre Sexualempfindung, which subsequently appeared in much enlarged and revised editions. It speedily superseded all previous books as a complete statement and judicious discussion of sexual inversion. Moll was not content merely to present fresh clinical material. He attacked the problem which had now become of primary importance: the nature and causes of sexual inversion. He discussed the phenomena as a psychologist even more than as a physician, bearing in mind the broader aspects of the problem, keenly critical of accepted opinions, but judiciously cautious in the statement of conclusions. He cleared away various ancient prejudices and superstitions which even Krafft-Ebing sometimes incautiously repeated. He accepted the generally received doctrine that the sexually inverted usually belong to families in which various nervous and mental disorders prevail, but he pointed out at the same time that it is not in all cases possible to prove that we are concerned with individuals possessing a hereditary neurotic taint. He also rejected any minute classification of sexual inverts, only recognizing psycho-sexual hermaphroditism and homosexuality. At the same time he cast doubt on the existence of acquired homosexuality, in a strict sense, except in occasional cases, and he pointed out that even when a normal heterosexual impulse appears at puberty, and a homosexual impulse later, it may still be the former that was acquired and the latter that was inborn. In America attention had been given to the phenomena at a fairly early period. Mention may be specially made of J. G. Kiernan and G. Frank Lydston, both of whom put forward convenient classifications of homosexual manifestations some thirty years ago. More recently (1911) an American writer, under the pseudonym of Xavier Mayne, privately printed an extensive work entitled The Intersexes: A History of Similisexualism as a Problem in Social Life, popularly written and compiled from many sources. This book, from a subjective and scarcely scientific standpoint, claims that homosexual relationships are natural, necessary, and legitimate. [122] [123] In England the first attempts to deal seriously, from the modern point of view, with the problem of homosexuality came late, and were either published privately or abroad. In 1883 John Addington Symonds privately printed his discussion of paiderastia in ancient Greece, under the title of A Problem in Greek Ethics, and in 1889-1890 he further wrote, and in 1891 privately printed, A Problem of Modern Ethics: Being an Enquiry into the Phenomena of Sexual Inversion. In 1886 Sir Richard Burton added to his translation of the Arabian Nights a Terminal Essay on the same subject. In 1894 Edward Carpenter privately printed in Manchester a pamphlet entitled Homogenic Love, in which he criticised various psychiatric views of inversion at that time current, and claimed that the laws of homosexual love are the same as those of heterosexual love, urging, however, that the former possesses a special aptitude to be exalted to a higher and more spiritual level of comradeship, so fulfilling a beneficent social function. More recently (1907) Edward Carpenter published a volume of papers on homosexuality and its problems, under the title of The Intermediate Sex, and later (1914) a more special study of the invert in early religion and in warfare, Intermediate Types among Primitive Folk. In 1896 the most comprehensive book so far written on the subject in England was published in French by Mr. André Raffalovich (in Lacassagne's Bibliothèque de Criminologie), Uranisme et Unisexualité. This book dealt chiefly with congenital inversion, publishing no new cases, but revealing a wide knowledge of the matter. Raffalovich put forward many just and sagacious reflections on the nature and treatment of inversion, and the attitude of society toward perverted sexuality. The historical portions of the book, which are of special interest, deal largely with the remarkable prevalence of inversion in England, neglected by previous investigators. Raffalovich, whose attitude is, on the whole, philosophical rather than scientific, regards congenital inversion as a large and inevitable factor in human life, but, taking the Catholic standpoint, he condemns all sexuality, either heterosexual or homosexual, and urges the invert to restrain the physical manifestations of his instinct and to aim at an ideal of chastity. On the whole, it may be said that the book is the work of a thinker who has reached his own results in his own way, and those results bear an imprint of originality and freedom from tradition. In recent years no one has so largely contributed to place our knowledge of sexual inversion on a broad and accurate basis as Dr. Magnus Hirschfeld of Berlin, who possesses an unequalled acquaintance with the phenomena of homosexuality in all their aspects. He has studied the matter exhaustively in Germany and to some extent in other countries also; he has received the histories of a thousand inverts; he is said to have met over ten thousand homosexual persons. As editor of the Jahrbuch für sexuelle Zwischenstufen, which he established in 1899, and author of various important monographs—more especially on transitional psychic and physical stages between masculinity and femininity—Hirschfeld had already contributed greatly to the progress of investigation in this field before the appearance in 1914 of his great work, Die Homosexualität des Mannes und des Weibes. This is not only the largest but the most precise, detailed, and comprehensive—even the most condensed—work which has yet appeared on the subject. It is, indeed, an encyclopedia of homosexuality. For such a task Hirschfeld had been prepared by many years of strenuous activity as a physician, an investigator, a medico-legal expert before the courts, and his position as president of the Wissenschaftlich-humanitären Komitee which is concerned with the defense of the interests of the homosexual in Germany. In Hirschfeld's book the pathological conception of inversion has entirely disappeared; homosexuality is regarded as primarily a biological phenomenon of universal extension, and secondarily as a social phenomenon of serious importance. There is no attempt to invent new theories; the main value of Hirschfeld's work lies, indeed, in the constant endeavor to keep close to definite facts. It is this quality which renders the book an indispensable source for all who seek enlightened and precise information on this question. Even the existence of such a treatise as this of Hirschfeld's is enough to show how rapidly the study of this subject has grown. A few years ago—for instance, when Dr. Paul Moreau wrote his Aberrations du Sens Génésique—sexual inversion was scarcely even a name. It was a loathsome and nameless vice, only to be touched with a pair of tongs, rapidly and with precautions. As it now presents itself, it is a psychological and medico-legal problem so full of interest that we need not fear to face it, and so full of grave social actuality that we are bound to face it. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Havelock Ellis (2004). Studies in the Psychology of Sex, Volume 2. Urbana, Illinois: Project Gutenberg. Retrieved October 2022 https://www.gutenberg.org/cache/epub/13611/pg13611-images.html This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org , located at https://www.gutenberg.org/policy/license.html.

THE STUDY OF SEXUAL INVERSION

On December 27th, 2022—just a couple of days after Christmas and less than a month post-ChatGPT’s grand reveal to the world—a team of Google and DeepMind researchers published on a new large language model they named and its in a medical and clinical setting. their findings MedPaLM “encouraging results” And while the globe is still firmly grasped in a ChatGPT frenzy, its virtually anonymous “cousin” MedPaLM could potentially herald a not-too-distant revolution in how we administer care. So What Exactly Is MedPaLM? To understand what MedPaLM is, let’s first break down what each letter in its abbreviated name means (a bit of an acrostic poem, if you will): Med: Medical MedPaLM is based on Google’s 540 billion parameters strong and was specially adapted and optimized to be tested on , a newly formulated benchmark for testing large language models (LLMs) for medical and clinical applications combining six existing open question-answering datasets spanning professional medical exams, research, and consumer queries (NedQA, MedMCQA, PubMedQA, LiveQA, MedicationQA, and MMLU) as well as a new dataset named HealthSearchQA focused on general medical knowledge searched for by consumers. PaLM model MultiMedQA Pa: Pathways Pa stands for , a new AI architecture launched by in October 2021, designed to handle a multitude of tasks simultaneously and learn new tasks at hyper-speed. Breaking away from the conventional method of training machine learning models to perform a single task well, Pathways ushered in a new era of AI models capable of performing multiple tasks at once while also drawing upon and combining their existing skills to learn new tasks faster and more effectively. Pathways Google - Jeff Dean, Google Senior Fellow and SVP, Google Research. “We want [this] model to have different capabilities that can be called upon as needed and stitched together to perform new, more complex tasks – a bit closer to the way the mammalian brain generalizes across tasks.” ‍ Google's Pathways AI architecture ‍ LM: Language Model LM is short for (in this instance, a Large Language Model), a computational model designed to understand and generate human language. Or, in simpler terms, a statistical model that predicts the probability of the next word in a given sequence of words. Language Model Given a sentence such as "The cat sat on the ____," a language model can predict that the most likely word to follow is "mat," "rug," or "chair," based on the probabilities of these words appearing in the context of the sentence. Large language models can be trained on vast amounts of text corpora using machine learning algorithms to learn the patterns and structure of language. Once trained, these models can be utilized for a range of natural language processing tasks, including text generation, translation, sentiment analysis, and more. How Was MedPaLM Developed? MedPaLM is the product of an extensive iterative experiment conducted by a team of Google and researchers to assess the efficiency and accuracy of large language models for medical and clinical applications. DeepMind ‍ What does that mean? Imagine you’re trying to perfectly tune a violin. To hit the perfect note, you need to continuously increase and decrease string tension and turn the knobs ever so gently to reach the desired outcome. In many ways, this is precisely what MedPaLM’s creators did to achieve their objective. So rather than thinking about it as an entirely new language model, think about it as Google’s PaLM model deliberately and meticulously calibrated for maximum precision in a medical setting. Ok, so how was it “tuned”? To tune or, more precisely, train MedPaLM, its researchers used prompts. Prompts for large language models are pieces of text used to guide or direct the generation of a new text by the model. They are intended to help the model produce output that is relevant to a particular task or topic and can take various forms, such as fill-in-the-blank statements, questions, or context-setting phrases. So, for example: "Complete the following sentence: The capital of France is ____." "What is the best way to prepare a grilled cheese sandwich?" "Write a paragraph describing the causes of the American Civil War." "In this passage from a novel, a character is introduced: 'She was tall and thin, with piercing blue eyes and a serious expression on her face.' What is the character's name and what is she doing in this scene?" Why do I need a healthcare-purposed language model to tell me the best way to prepare a grilled cheese sandwich? You don’t. This is exactly why MedPaLM’s developers employed a technique called Instruction Prompt Tuning (IPT).  Instruction Prompt Tuning is a method of fine-tuning a large language model for a specific task by providing examples of desired input and output pairs in the form of prompts. So rather than construct an LLM with general answers to general topics, MedPaLM’s team used guidelines and exemplars from a panel of qualified clinicians for each consumer medical question-answering datasets to prompt tune Flan-PaLM (MedPaLM’s previous iteration) and form a hyper-healthcare-focused LLM: MedPaLM. ‍ ‍ So How’s MedPaLM Performing? Well, that really depends on what you want it to do. In terms of providing answers to consumer medical questions, 94.4% of MedPaLM’s answers were judged as directly addressing the user’s question intent—only 1.5% less than a human clinician. 80.3% of its answers were found to be helpful, 11% less than a human clinician but 30% better than its predecessor Flan-PaLM. Impressive? You bet. ‍ ‍ But in other fields tested on, MedPaLM leaves much to be desired. In 16.9% of its answers, MedPaLM retrieved incorrect information as opposed to 3.6% for human clinicians. Incorrect reasoning was seen in 10.1% of MedPaLM’s answers and in just 2.1% of clinician answers. Incorrect comprehension occurred in 18.7% of cases for MedPaLM and in 2.2% for clinicians. What Does This Mean for Healthcare? Trying to definitively predict how MedPaLM and other LLMs will be harnessed for clinical and medical purposes is a bit of a fool's game at this stage. As impressive as recent advancements in this field are, a single error could have disastrous results within the life-or-death stakes of medical care. One incorrect answer to a patient's query (i.e., what drug should be taken to address a specific condition) could lead to hospitalization and even death. This issue is further exacerbated by the fact that large language models are essentially enormous black boxes and lack the explainability or the ability to pinpoint exactly why they gave a certain input. Solving an error starts with knowing how to find the problem, and as spectacularly capable as MedPaLM may be, that's just one (out of many) things it can't do. So, What Can LLMs be Used for in Healthcare? Approximately 7,287 articles ( ) are published monthly in medical journals. To theoretically keep up with the literature, a physician would need to dedicate 627.5 hours per month, or about 29 hours per weekday, or 3.6 full-time equivalents of physician effort to the task. And though this isn't realistic in any scenario, LLMs are incredibly effective in summarizing, analyzing, and presenting key points from long-form text, drastically cutting down the time needed to stay up to date with the latest research. Summarizing and analyzing research papers: with an average of 5.16 minutes reading time Large language models can be used to improve the quality of clinical documentation by identifying missing or incomplete information in medical records. Improving clinical documentation: By generating appropriate codes for procedures and treatments, LLMs could help reduce errors and ensure clinicians are adequately compensated for their work. Coding and billing: Using LLMs, clinicians can identify trends and patterns in patient health and develop targeted interventions to improve population health outcomes. Population health management: By analyzing large amounts of data and identifying patterns and trends that may not be easily traceable with traditional statistical methods, LLMs can be employed to facilitate clinical research at increased velocity and scale. Clinical research: LLMs can be leveraged to sift through and analyze patient feedback and sentiment through thousands of surveys and social media interactions to identify critical areas for improvement in patient satisfaction, access, and experience. Patient satisfaction analysis: What About Appointment Scheduling? LLMs are only as good as the data they’ve been trained on. ChatGPT, for example—albeit its objectively impressive knowledge retrieval capabilities—was only trained on parameters dating to 2021. Any event that took place at any time past its cut-off date of September 2021 is entirely out of its scope. So even though Boris Johnson resigned as Prime Minister of the United Kingdom in September 2022 and has since been succeeded by two prime ministers (Liz Truss and currently Rishi Sunak), ChatGPT is (self-admittedly) completely unaware. ‍ ‍ But what if I train an LLM on a list of all of my health system’s physicians? Wouldn't that work? It would. But only until the moment even the most minute change takes place. So, for instance, if just one of your physicians moves to a different clinic, adds another specialty to their care offerings, updates their working hours, or even resigns, your appointment-scheduling LLM will provide your patients with wrong and misleading information. If even a single piece of data is no longer relevant, the entire model becomes corrupted and, therefore, obsolete. The same can be said for any use case requiring real-time data to perform accurately. So, for example, site search, prescription refills, and up-to-date FAQs. ‍ ‍ Oh, so LLMs can really only help me with administrative tasks? On their own, yes, which is nothing to sneeze at and will save healthcare practitioners innumerable hours otherwise spent tackling bureaucratic office work. But for any use cases that demand real-time, up-to-date data, LLMs can only serve as an ingredient of a much more adaptive and flexible conversational platform, not the entire dish. , Hyro’s CEO lays out the ingredients needed to mix LLMs with enterprise-level conversational AI platforms to create truly elastic patient interactions that are driven by continuously updated organizational knowledge (proprietary data) and dynamic business logic (ever-changing internal and external processes). In this in-depth analysis Israel Krush Big Hero 6 When asked in a recent what he thought of ChatGPT and other LLMs, Andy Chu, SVP of Product and Technology at Providence Digital Innovation Group, said the industry is still a a 2014 Disney film depicting a futuristic healthcare robot—But added he envisions this technology used to answer patient-related administrative, "decision-tree", or general health-education questions. Becker's Hospital Review interview "long way off from the days of Big Hero 6,” https://youtu.be/jbKSvOfqxF0?embedable=true Chu's thoughts echo a broader sentiment shared by most digital health leaders that LLMs such as MedPaLM should be greeted with equal parts of excitement and caution. While there's no question these extraordinary "machines" can automate and accelerate a vast spectrum of primarily administrative applications, there are still glaring red flags when it comes to medical and clinical use. Furthermore, LLMs' absence of data adaptability renders them ineffective for any goal which calls for current and evolving internal organizational knowledge. Chu put it best by saying: "It's important to remember that healthcare is very personal, and generative AI technologies are as good as the data accessed. Data sources within healthcare are rightfully secure and protected, so we must be careful not to overhype the promise — especially when it comes to areas such as symptom detections, triage, and clinical treatment suggestions." Also published here.

An Intro to MedPaLM: ChatGPT's Healthcare-Focused "Cousin" 

Attention Deficit Hyperactivity Disorder, or ADHD, is a neurological disorder that struggles with. around 2.8% of the world population It is caused by a deficiency in neurotransmitters that affects chemicals like dopamine, resulting in impulsivity, attention deficit, hyperactivity, and difficulties with emotional regulation. ADHD symptoms can appear as early as 3 to 6 years old and persist throughout one’s lifetime. Extensive research on ADHD started less than two decades ago and hasn’t yet led to any significant breakthroughs. Medical professionals believe that there are several underlying symptoms and indicators of ADHD that are still unidentified, and researchers need a lot of data to conduct a viable study. Recently, have poised gaming as an opportunity to advance ADHD research. several papers As the titles become more engaging and immersive, not just solely narrative-driven or focused on problem-solving, players demonstrate their ability to focus and show emotion while their gameplay style reflects their unique psychological traits. Hence, these observations and collection of behavioral Big Data can provide critical insights for new ADHD studies, helping researchers get a better understanding of this disorder. Gaming: A Source of Precious Data? According to studies published in the and journal, observing individual behavior in gaming can generate essential data for ADHD diagnosis and identifying new symptoms. Journal of Attention Disorders European Child Psychiatry While people with ADHD symptoms have difficulties focusing, have shown that they find video games, especially action ones, appealing due to their fast-paced, interactive nature. some studies Through using wearables and brain-computer interfaces, researchers can match elements such as chosen strategy, reaction speed, and brain response with physical parameters like age and blood pressure to identify symptoms proactively. Currently, a few titles are being designed to assess different aspects of ADHD, such as attention, impulse control, and working memory. For instance, the game ’Project: EVO’ was explicitly developed for tracking symptoms through players’ performance. this game as a prescription for early-age ADHD. The FDA even recommends Through this method, data can be collected in a less biased way. Traditional assessment methods, such as questionnaires and cognitive tests, rely on self-reporting, which can be less reliable. Video games, on the contrary, can track and record the player's performance in real time, providing researchers with more accurate and objective data. The fun and engaging nature of gaming can make the research process more enjoyable for participants. This is particularly important when working with children who may find traditional assessment methods boring and unengaging. The Combination of AI and Gaming Can Be Revolutionary The combination of human care and AI (artificial intelligence) analyses based on Big Data will help increase the accuracy of one’s treatment scenario, as it did with , such as early knee osteoarthritis detection. other disorders and illnesses Gaming and AI can be used to develop personalized treatment plans tailored to the specific needs of individuals with ADHD. By analyzing their behavior and cognitive patterns, AI algorithms can help clinicians identify areas needing improvement and develop customized treatment plans incorporating gaming activities to improve their focus and other cognitive abilities. Framed as an enjoyable activity rather than a tedious test, video games provide a new and exciting way of studying ADHD. They offer a level of engagement and stimulation that traditional assessment methods lack, providing researchers with real-time and objective data. With the development of more games specifically designed for this purpose, we expect to see a significant increase in the use of gaming as a tool for studying this disorder.

Stories, News & Mentions about CLINICIA

Clinicia’s Stories on HackerNoon