At Neural Labs Africa Innovation Hub we use Deep Learning and Computer Vision to identify diseases in Real time. Our AI algorithm NeuralSight™ is capable of identify over 20 respiratory, heart and breast pathologies which include: Pneumothorax, Cardiomegaly, Benign breast Tumor, Malignant ...

Neural Labs Africa

Neural Labs Africa is using AI-enabled medical imaging (NeuralSight) for real-time diagnosis.

Neural Labs Africa - Crunchbase Company Profile & Funding

#HealthTech #AIinHealthcare #NeuralLabsAfrica #HealthcareInnovation #AfricaHealthcare ... Portfolio Strategy, Venture Fund (AI and Data Science + Climate Cohorts), UNICEF Global Office of Innovation 3mo · Recently How we made it in Africa published an interview with Villgro Africa CEO Wilfred Njagi about investment opportunities in the healthcare sector in #Africa. The example of UNICEF Innovation Venture Fund Alum Neural Labs Africa -- a company ...

Neural Labs Africa | LinkedIn

Neural Labs Africa has develped solution that uses computer vision for diganosing in real time over 20 respiratory, heart and breast diseases,... ... Find out more about Neural Labs Africa including the VentureRadar Innovation and Growth scores, Similar Companies and more.

Neural Labs Africa | VentureRadar

Information on valuation, funding, cap tables, investors, and executives for Neural Labs Africa. Use the PitchBook Platform to explore the full profile.

Neural Labs Africa Company Profile: Valuation & Investors | PitchBook

Neural Labs Africa Ltd is an innovative medical technology company using Artificial Intelligence (AI) to transform medical imaging diagnosis. Their mission is to positively impact healthcare by using AI, a revolutionary technology to democratize access to diagnostic healthcare.

Neural Labs Africa | UNICEF Venture Fund

Neural Labs Africa - AI-based medical imaging and diagnosis solutions.

Neural Labs Africa - Company Profile - Tracxn

Neural Sense is a neuromarketing company. We conduct neuroscience driven market research using Brain Monitoring (EEG; fNIRS), Eye-tracking &amp; Biometric technologies to better understand, affect and elevate customer experiences. Our services include Advertising Testing; User Testing and Shopper ...

Neural Sense Neuromarketing | Neuroscience Driven Marketing Research, ...

Neural Labs Africa | UNICEF Innovation Fund

LPR software for license plate, container id and all the actors in the streets recognition. Traffic analysis and access control solutions.

Neurallabs- Video analysis for Smart cities and ITS

Neural Labs Africa Ltd - Introduction Neural Labs Africa is an innovative medical technology Company using Artificial Intelligence (AI) to transform medical imaging diagnosis.

About Neural Labs Africa Ltd - Software Development company in ...

Passionate about the vibrant tech startups scene in Africa, Tom can usually be found sniffing out the continent&#x27;s most exciting new companies and entrepreneurs, funding rounds and any other developments within the growing ecosystem.

Neural Labs - Disrupt Africa

Company types: Contractor / building / roadbuilding /earth-moving, Consultancy firm, Manufacturer, Public transportation, Government, local (municipality), Government, central (Ministries), Government, provincial, Police, Transport business

Company profile

An innovative medical technology company screening medical images for radiologists and hospitals in real time to identify common indicators of tuberculosis …

Neural Labs Africa Ltd – VC4A

Helping industries transition to AI &quot;&gt;&quot;&gt;Empowering business with data-driven Al solutions for performance optimisation Contact us AI for Aviation Automate airport operations and allocates resources efficiently through data-driven decision-making to provide future-ready passenger experienceAviation ...

Home - Neural Lab

Twitter

Neural Labs Africais in the Healthcare IT Industry

157 Stories To Learn About Meet The Writer

237 Stories To Learn About Innovation

Security

The Age of Exciting Opportunities This article will transform the living conditions of millions of disabled people around the world who actually have the financial wherewithal to use these modern technologies. How Technology is Making the Impossible Possible “The blind see, the deaf hear, the lame walk again and the dumb speak.” - Matthew 11:5 No, I’m not getting all-spiritual on you! What I am saying is that today, technology, artificial intelligence, machine learning, human-computer interfacing, bio-implants, cybernetics, and bio-prostheses have combined to help – literally - “The blind see, the deaf hear, the lame walk and the dumb speak.” So how do I show you these remarkable achievements? I thought, let’s go practical. We are going to use four real-life innovations – modern technology, using which the blind are enabled to see (well, more like audio assistance in most cases), the deaf are given hearing, the lame are given mobility, and the dumb are given speech. I haven’t been in a very long time! this excited Let’s get started! The Blind Will See There are two sides to this coin. With the current state of the art technology, thee are two ways in which the blind will ‘see’. The first has more to do with actual sight – bionic eyes. (From https://circuitdigest.com/article/bionic-eye ) A Bionic Eye is a technology fundamentally built around computer vision, human-computer interfacing, neural circuitry implants, and the receptor sensors themselves. Basically, they allow the light waves entering the bionic eye to be processed in a manner similar to the ways natural eyes work themselves. This is a considerable feat of engineering, but the human eye is such a complex and advanced entity that modelling it is a formidable challenge. However, these bionic eyes have restored basic eyesight to those suffering from illnesses such as retinal pigmentosa and macular degeneration. The eyesight restored is very primitive. However, it is a massive leap from seeing nothing at all, and technology will only improve this state-of-the-art device as time progresses. An AI Assistant to the Blind What if I told you that there was already an app installable on your phone that describes everything your phone camera points to? With the right configuration, this will change everything as far as a blind person is concerned. ‘ A blind person, with this app, that uses image recognition, computer vision, and deep neural networks has, for all purposes, a digital ‘friend’ who can describe everything around them, be it the denomination of currency notes or the expression of a persons face. For more, see below. EyeSense Is A New AI App That Helps the Blind Discover the World ID Labs developed EyeSense, an Artificial Intelligence application that helps blind and visually impaired people discover their surroundings, while also enhancing human-to-human connection. The AI app is also of great value for individuals with mild cognitive impairment – including memory, language, and thinking skill. EyeSense uses a deep learning-based computer vision to recognize objects and special facial features – like smiles and winks – with no need for an internet connection. Just point the EyeSense camera towards an object of interest and listen to the app identify it. EyeSense has the ability to easily “learn” new objects and recognize them from many different viewpoints in real-time. This can be done with no internet connection and with an easy training process. The design philosophy behind EyeSense lies in the ability to personalize each user’s expersience, and to enhance human-to-human interaction. is thus a remarkable technology-assisted enabler for the blind, for two simple reasons EyeSense it works offline it is capable of user personalization. Human-Computer Interfaces are going to change the ways we think, live, and work. All that you’ve seen till now is just the beginning. A revolution will take place, giving humans computer-like abilities and computers human-like abilities. It’s a very exciting prospect! Home Automation – the Role of Google Home and Amazon Echo AI-devices that manage homes like Amazon Echo provide unprecedented access to the entire world for the visually impaired. With Google Home, for example, one can surf the net, listen to songs, send messages, set alarms, start and stop home appliances, access radio channels. And much, much more. That too, the cheapest of these devices, the Amazon Echo Dot, is just 49 USD! The world is changing. Don’t let a blind person remain ignorant of these technological advances and And this home automation task list is just growing bigger and bigger every single minute, every single day. educate, educate, educate. The Deaf Will Hear. Now this has some precedents. Blindness, of course, is the greatest disability. But being deaf is bad as well! How can AI possibly improve the lives of the deaf? As it turns out, it can. And massively so! Speech Recognition, Captioning, and Conversion of ASL to Spoken English Microsoft has partnered with several industry leaders to bring a audio-enabled classroom to several universities in the United States and abroad. Perhaps Microsoft’s best move was to appoint a wonderful leader and a deaf person herself, Jenny Lay-Flurrie, who has brought in some wonderful innovation to the deaf using speech recognition technology and automatic speech captioning. This makes it possible for the deaf to hear what people are saying with apps that translate multiple languages to English text that people can read on the phone or their laptops. In fact, the technology is even capable of converting American Sign Language (ASL) to English in real-time! Bionic Ears You might say, all right, Thomas. AI translates speech to visual text. But you promised that the deaf would hear. Actually hear! Yes - I did. Have a look at one of the latest discoveries in hearing aids – bionic ears. The quote from http://www.bionicsinstitute.org/stories/bionic-ears-and-how-they-change-the-brain/: 30 years ago, the potential for deaf children to communicate verbally was greatly diminished. Now with bionic ears, and educational support, people with even a profound hearing loss can speak clearly and confidently – many even learn multiple languages. 5-year-old bilateral cochlear implant recipient Alana Brown has a passion for music and dancing. Alana was born profoundly deaf and received her first cochlear implant at 16 months and her second at 22 months. Alana’s family have written a poem about the change this made in their lives We cried the day that we found out, she could not hear a whisper even a shout. Sounds of music and birds that sing she couldn’t hear a single thing. Now a miracle has made her hear, it bought us hope and joy and many more tears. Now our girl can hear these things, can dance and move and even sing Doctors and audiologists observed that children implanted early in life like Alana respond very well to communications training. This was thought to be the result of brain plasticity – the ability of the brain to re-organize itself by forming new connections between brain cells (neurons) via new experiences. In the case of the bionic ear, the new experience is sound, from the device stimulating the auditory nerve in the cochlea. And this is not even examining the many and numerous advances that have taken place in the field of hearing aids assisted technologies. Is anyone you know deaf? Well – The only barrier between them and hearing is finance. Nothing else! Now you may have a question – how are all these advances taking place? What is the underlying technology behind all these developments? The answer is or for short. What is so special about them that enables them to do all these seemingly impossible things? I promise to devote one section to that at the end of this article. For now, let’s move on! artificial neural networks ANNs The Lame Will Walk This is one subject that is – well – easy to analyse. Being lame is one of the most serious disabilities a person can have. There are varying degrees of lameness. But to show how AI can contribute to overcoming this obstacle – well – have a look below! Bio-Prostheses How can AI help the lame to walk? The answer in one word – robotics! Well, to be more accurate – Human-Computer Interfacing through electrical chips and robot appendages wherever the missing limbs may be. There is even a video of a girl actually playing violin with a robotic arm, I won’t embed it but I am giving the link below just so that you, the reader, will understand exactly the amazing possibilities for someone missing any part of their body to enable normal functioning using technology and AI (the fuzzy logic control system for each limb, leg, arm, hand, foot, etc, used instead of PIDs (Google it), yes, that’s AI too!). The Miracle Violinist That Had Judges in Tears - Manami Ito’s World’s Best Audition Prostheses are essentially synthetic duplicates used in place of your missing limb. To connect a limb to the brain, so that the handicapped patient can use it properly, a little computer circuitry is need, as well as a knowledge of the brain. The wonder between our ears that all human beings on Earth possess is essentially an electromagnetic machine, with really, really, low voltages. Sending signals to and from the brain is possible using intricate bio-electric circuitry. How does this control system work? You guessed it – (in most cases). artificial neural networks Don’t confuse ANNs with real biological neurons, that equipment interfacing with the brain’s electromagnetic signals is a piece of computer circuitry – an embedded microprocessor. The control system that maps the brains signals to the movements of the prosthesis – which is a pattern recognition problem – is where you will find the ANN. Brain neurons have computer circuits and chips that are implanted so that the interfacing works correctly. The Dumb will Speak Now this is one of the most well-known and well-documented cases where those incapable of speech were given the ability to communicate through computers and technology. We will take one of the most celebrated of the dumb and the paralyzed – the cosmologist Stephen Hawking. Stephen William Hawking was born in 1942 and was a theoretical physicist. In 1963, he was diagnosed with progressive motor neuron disease (MND; also known as or Lou Gehrig's disease) that gradually paralysed him over the decades. amyotrophic lateral sclerosis "ALS" Even after the loss of his speech, he was still able to communicate through a speech-generating device, initially through use of a hand-held switch, and eventually by using a single cheek muscle. He died on 14 March 2018 at the age of 76, after living with the disease for more than 50 years. What is so remarkable about Stephen Hawking was his incredible prolificity despite his illness. Even while completely paralysed and unable to move, let alone talk, he published several books, authored an innumerable number of research papers, the last being published in 2018 just before his death, changed the entire world’s understanding about life and history, and became one of the most iconic of all British celebrities. His most famous book, sold over 9 million copies and was translated into several languages. A Brief History of Time How did Hawking Communicate? As the disease spread, Hawking became less mobile and began using a wheelchair, and finally was totally paralysed barring one muscle in his left cheek. A speech-generating device combined with a software program, served as his electronic voice, allowing Hawking to select his words by moving the muscles in his cheek. Think about it. He supervised 39 PhD students, won practically every single award available for scientists worldwide, with an illness so crippling that he was given a life expectancy of 2 years in Once again, it was and a speech generator using cheek gestures (another victory for the artificial neural network construct) that helped him speak, and it was developed at the University of Cambridge. 1963. AI to the rescue The Artificial Neural Network Magic Here, we’re going to take a little detour and deliver the goods on neural networks as promised. Why are neural nets so popular and so widely used? The answer lies in an AI science concept called the Universal Approximation theorem. to quote Wikipedia: In the mathematical theory of artificial neural networks, the universal approximation theorem states that a feed-forward network with a single hidden layer containing a finite number of neurons can approximate continuous functions on compact subsets of Rn, under mild assumptions on the activation function. The theorem thus states that simple neural networks can represent a wide variety of interesting functions when given appropriate parameters; however, it does not touch upon the algorithmic learnability of those parameters. One of the first versions of the theorem was proved by George Cybenko in 1989 for sigmoid activation functions. Kurt Hornik showed in 1991 that it is not the specific choice of the activation function, but rather the multilayer feedforward architecture itself which gives neural networks the potential of being universal approximators. The output units are always assumed to be linear. Let me translate that into English for you. Given enough data, a neural network with at least one hidden layer can map any real number set to any real number set. What does that mean? It means that you can ‘approximate’ any process in the world and extrapolate, predict, interpret, interpolate, forecast and simulate any natural process using a sufficient amount of data. But that opens up the whole world! Every single process is a function of some sort or the other! A process with inputs and outputs can be viewed as a function. What do we know about it? Nothing! But that’s the beauty of it – we don’t need to know anything about the function! Given enough input and output data, a neural network can approximate practically anything you want in the entire universe. To help you understand, here is a woefully incomplete list of the applications of artificial neural networks: Speech Recognition. Pattern Recognition Pattern Detection Image Processing Self-Driving Cars Speech Processing (Stephen Hawking?) Natural Language Processing Forecasting Regression Classification Fuzzy Control Systems (artificial leg?) Cancer Detection Anomaly Detection Video Games Playing AI Deep Learning Reinforcement Learning Scientific Simulation Voice Recognition Face Recognition … I can easily add 30 more entries without losing my breath, unfortunately I’m already over my word limit. So, I hope you understand now. Artificial Neural Networks act as Universal Function Approximators. As simple as that. Call for Compassion If there is one word to sum up this entire 21st century, it is indifference. Syrians crucified in the Middle East. People dying of hunger, thirst, and famine in Africa. Somalia. Ethiopia. North Indians butchering fellow countrymen over a cow. And that same cow treated with more respect than women are treated with in India (famously titled Hindustan/Rapistan). What’s our reaction? (My words – from the lips of Thomas Cherickal) “What’s there for breakfast today? I just finished reading the paper.” We need to identify and act for the oppressed. Sponsoring AI devices for the poor who can’t afford it all over the world would be an incredible first step. Any millionaires reading this? Mark Zuckerberg? Bill and Melinda Gates? (yes, their foundation already does wonderful work). Mukesh Ambani? Amir Khan? Roger Federer? Sachin Tendulkar? Priyanka Chopra? I know all of us want to leave a good impact on the world. A lasting legacy. So, my plea to all the super-rich and super-famous – be role models. Give technology that removes disabilities to the underprivileged (in person - so that you’re not scammed). Make it a mega-project. As for me, I’m going to see how many months I need to earn enough to sponsor a prosthetic for a lame child in Africa. Be human. Be genuine. Be compassionate. “ ”― Mandy Hale, The Single Woman: Life, Love, and a Dash of Sass There is nothing more beautiful than someone who goes out of their way to make life beautiful for others. Cheers! References Blindness https://www.huffingtonpost.com.au/2017/11/15/artificial-intelligence-is-helping-blind-people-to-see_a_23278974/ https://circuitdigest.com/article/bionic-eye http://id-labs.org/ai-app-eyesense/ https://www.visionaware.org/info/everyday-living/helpful-products/using-apps/seeing-ai-app/1235 Deafness https://blogs.microsoft.com/ai/ai-powered-captioning/ http://www.bionicsinstitute.org/stories/bionic-ears-and-how-they-change-the-brain/ Lameness https://en.wikipedia.org/wiki/Prosthesis Dumbness http://www.hawking.org.uk/ https://en.wikipedia.org/wiki/Stephen_Hawking https://www.scienceabc.com/innovation/stephen-hawking-cheek-communication-help-computer-speech-generating-device.html

How Emerging Tech Will Revolutionize The Life of Physically Challenged People

298 Stories To Learn About Facebook

The Chinese Social Network

My pandemic project was . As everyone knows by now, we need to sequence the genetic code of pathogens like SARS-CoV-2 to identify and track their most dangerous variants. In the first months of the pandemic, most developing countries did not have the capacity to sequence the virus. AccessibleGenomics.org We built a global team of volunteers, raised some funding from , got donations from suppliers, and organized online training for our partner lab, PGC Mindanao. 10 months after we started, that lab made history: the first sequencing of SARS-CoV-2 enabled by volunteers, and the first on-site sequencing of any organism in Mindanao, Philippines. Just One Giant Lab I shared my project management lessons learned . At the time of writing, we are interviewing labs in Africa and Latin America to see how we can best share what we learned from enabling a rural lab in a developing country to start sequencing. in this webinar On the positive side, I was surprised and inspired by all the help we got from the global genomics and open science communities. On the negative side, I was surprised by what turned out to be the biggest threat to the project. It was not the difficulty of learning sequencing (turned out to be easy for experienced wet lab folks) or getting funding (hard but not impossible). What almost killed the project was the supply chain. Sequencing reagents need to be imported from the US or Europe. This makes them expensive for a lab in the global south. On top of that, we learned that you need to add 30% to 100% to pay gate-keepers, both legal and corrupt ones. Since we grew up with open-source alternatives, we naturally asked if there are open-source versions of the reagents—affordable ones that allow us to bypass gate-keepers. Not yet, it turns out, but it looks like we are almost there. The answer lies in synthetic biology, a field that is at the cusp of its own open-source revolution. What is synthetic biology and how will it make our lives better? Most things that keep us alive are made of atoms strung together to make molecules. Back in our chemistry lab classes, we strung together atoms through simple chemical reactions, resulting in products with characteristics different from their reactants. There are some valuable materials that are too complex to be made in a chemistry lab, but which we take for granted as products of living organisms. For instance, the food we eat, the air we breath and the medicines we take. One of the biggest discoveries in the last century was figuring out how living organisms create these complex molecules. We eventually figured out how to engineer this process, particularly in microorganisms, like the bacteria . This allowed us to make it produce life-saving complex molecules like insulin. E. coli I have two uncles who need to inject themselves with synthesized insulin. I'm sure you also know of people whose lives depend on insulin. Today, approximately 100 million people around the world need insulin. Producing insulin at this scale is a result of synthetic biology. If you are in IT, you'd know about Unix and its open-source counterpart, Linux. You'd probably also know that over 90% of the world's top 1 million web servers run on Linux. If you are an open-source aficionado, you'd know about Red Hat, a competitor to the likes of HP, IBM, and Oracle that built a huge services business supporting open source instead of proprietary software. If you are in the field of synthetic biology, or "synbio", you could probably give examples of synbio counterparts of Unix, and of companies like HP and IBM. However, you'll probably be hard-pressed at giving examples of their open-source equivalents in the synbio world. That's because we're still in synbio's pre-open-source phase. Let's look at previous open-source revolutions to see what might be needed to create open-source synthetic biology. How did the Open Source revolution happen? I'm grateful to my parents for getting me and my siblings an encyclopedia in the house where we grew up. Those hardbound books were stacked taller than I was and cost a lot of money. The money my parents paid went to a global business that included expert writers, editors, project managers, printers, and door-to-door salespeople. If you'd told me that volunteers will build an encyclopedia better than that global business, I would not have believed you. But this is exactly what happened. Wikipedia covers a lot more topics with greater depth than any of those encyclopedias made by hired professionals. It turns out people are wired to give freely to the community once they are past survival mode. And it turns out Wikipedia was just following the footsteps of the trailblazer, open-source software. Both Wikipedia and open-source software were enabled by the following: An open community collaboratively creating a resource free for the world's use A legal framework that protects the freedom of that community's creations Broad access to equipment to contribute, in creating that free resource An open community collaboratively creating a resource free for the world's use I was a Wikipedia contributor back in 2005 - 2010. I was a combatant in a number of edit wars. I sometimes still visit talk pages to see the story behind a Wikipedia article. Alliances, rivalries, and even friendships were formed in those talk pages. Whatever our viewpoint on a topic was, we were united by a common goal and a common culture. We wanted well-written articles on topics we cared about. And we tried to abide by the principles of writing from a neutral point of view and assuming the good faith of fellow contributors. Open-source software projects are also fueled by open communities with a common goal and rules of engagement. A legal framework that protects the freedom of that community's creations You'd want the law to be your ally if you want it to protect, or at least to not attack, the goods your community creates. Copyright law was originally intended to protect creators from copycats. Book authors and software developers also need to feed their kids. The law was intended to make sure they are paid for their work. What if your kids are already taken care of, and you just want to give away your work? This was a novel situation in the past decades. GNU GPL, Copyleft and Creative Commons were solutions within copyright law that protected not the income of creators, but the freedom of their creations. Broad access to equipment to contribute to creating that free resource Prior to the internet, the ability to contribute to projects like Wikipedia or open-source software like Linux was limited to academics in rich countries. Because of the lowering cost of computation, hardware, and software, those of us in the periphery can now contribute to these projects. Synbio is entering its open-source phase To see the parallels between software and synbio, let's first look at an example of a company built on proprietary synthetic biology, the counterpart of 80s' HP and IBM. is one of the unicorns in the synbio world. They engineer microorganisms for your business, so you could produce valuable molecules. For instance, Ginkgo is engineering for Cronos Group, a cannabis company, to produce cannabinoids without plants. Ginkgo Bioworks E. coli Their technology is proprietary, so it is a secret. But we know that they certainly have a wetware stack, a hardware stack, and a software stack. In synbio, "wetware" is the biological materials that you use for genetic engineering and fabrication, like engineered organisms and chemical reagents. Hardware is the machines that enable the same processes, like liquid handlers and bioreactors. And software are applications that help design these organisms and run the hardware. The collection of these stacks is called a biofoundry. Open-source software helped democratize access to computing. Similarly, an open-source biofoundry will democratize access to synthetic biology. I'm optimistic. We could already see in synbio the three ingredients that we saw in Wikipedia and in open-source software. First, an active, collaborative community has formed around synbio to develop and share ideas, resources, and technologies. For instance, the annual and global genetic engineering student competition, , has been seeding the open-source synthetic biology community for years. looks like the synbio counterpart of Y Combinator companies. For example, iGEM alumni are behind Ginkgo Bioworks as well as Just One Giant Lab, the open science platform that funded Project Accessible Genomics. iGEM This list of companies built by iGEM alumni Second, legal frameworks analogous to GNU GPL and Copyleft have been created for wetware. For instance, wetware under can be legally used both by academic institutions and by third-world entrepreneurs like me who want to produce reagents locally, to lower their costs and bypass rent-seeking or corrupt gate-keepers. Organizations like are using open source to solve the insulin pricing problem in the US. , , and are examples of organizations that distribute or promote free wetware. OpenMTA Open Insulin Freegenes Public Domain Gazette Reclone Finally, we are seeing in synbio a parallel to the evolution from expensive mainframes to personal computers that democratized contributing to open-source software. For instance, Project Accessible Genomics was only possible because of nanopore sequencing. Just a few years ago, a sequencer would cost more than a car. Today, you can get a MinION sequencer for less than the price of an iPhone. Opentrons has dropped the price of automated pipetting. Bentolab and miniPCR have done the same for core molecular biology lab equipment. (Opentrons and Bentolab are also built by iGEM alumni.) We are headed in the right direction, but there is still a lot of work to be done. Let's democratize synthetic biology How would a low-cost biofoundry look like? This just happens to be the goal of a 2021 iGEM team called . Like Project Accessible Genomics, the team also got funding from Just One Giant Lab. I've been helping out the team through project management training. Friendzymes Here's the intro video of the team. They are working on multiple innovations to build a frugal biofoundry. The two I'm most excited about are the following: E. coli is the workhorse of synthetic biology. But using it tends to be expensive because you need to literally extract the product out of its guts (purification). What if you could engineer organisms that secrete the product from their cells naturally? The team is looking at and as alternatives to . Bacillus subtilis Pichia pastoris E. coli The lab skills you need to do synthetic biology require a lot of experience and wasted reagents. What if you use robots instead, like Opentrons? Then you could do your experiments and processes at scale and share them with other labs, the way 3D printer designs are shared now. I'm excited about the possibilities that open-source synthetic biology will open for us here in the developing world. It will make molecular biology research much more affordable for our scientists. It will also have an impact on medicine, pathogen biosurveillance and food security. Let's help Friendzymes by spreading the message of open-source synthetic biology or contributing to their (tax deductible for US donors)! crowdfunding campaign Thanks to Sam Bhagwat and Dr. Nicole Wheeler for reading and giving feedback on early versions of this article.

How Open-Source Synthetic Biology Can Make Medical Solutions More Accessible

Africa has over languages however, these languages are not well represented in the existing Natural language processing (NLP) ecosystem. One of the challenges is the lack of useful that can be used to solve different social and economical problems. 2000 African language datasets In this article, I have compiled a list of African language datasets from across the web. These datasets can be used in numerous NLP tasks such as text classification, named entity recognition, machine translation, sentiment analysis, speech recognition, and topic modeling. This collection of datasets have been made public to give you an opportunity to use your skills and help solving different challenges. Text Classification Text classification datasets are categorized or organized into different groups based on their contents. Below is the list of African language datasets for . Text classification 1. Swahili news Dataset The Swahili news dataset contains more than from different news categories such as Local, International, Business or Financial, health, sports, and Entertainment. The Swahili language is one of the most spoken languages in Africa, it is spoken by 100-150 million people across East Africa. 31,000 news articles The data was collected from different news publication platforms inside and outside of Tanzania. The dataset can be used to develop a multi-class classification model to classify news content according to their specific categories specified. The model can be used by Swahili online news platforms to automatically group news according to their categories and help readers find the specific news they want to read. You can also download this dataset from the : datasets python library datasets load_dataset

dataset = load_dataset( ) from import "swahili_news" The Swahili news dataset has an imbalance of category distribution. It contains few news articles in the following categories: Note: International News( 6.2%) Health News(4.9%) Business News(4.3%) 2. Chichewa News Dataset This dataset consists of in Chichewa. Chichewa is a Bantu language spoken in much of Southern, Southeast, and East Africa, namely the countries of Malawi and Zambia, where it is an official language. news articles The dataset contains a collection of containing over , and over . The Chichewa news articles have been categorized into such as education, law/order.politics, culture, arts and crafts, farming, economy, and wildlife. 3,482 articles, 930,000 words 48,000 sentences 19 categories You can also download this dataset from the following link: . AI4D Malawi News Classification Zindi Challenge Named-entity Recognition Named-entity Recognition datasets are used to extract information by locating and classifying named entities mentioned in unstructured text. Examples of entities are person names, organizations, locations, times, and dates. NER is an essential component of numerous applications including spellcheckers, conversational agents, and localization of voice and dialogue systems. Below is the list of African language datasets for Named-entity Recognition. 3.Masakhane-ner Datasets Masakhane is a grassroots NLP community for Africa, by Africans with a mission to strengthen and spur NLP research in African languages. The community created the first large publicly available high-quality dataset for named entity recognition (NER) in ten African languages. Amharic Hausa Igbo Kinyarwanda Luganda Luo Naija Pidgin Swahili Wolof Yorùbá You can read the research paper here and download the ten NER datasets . MasakhaNER: Named Entity Recognition for African Languages here Machine Translation Machine translation (MT) is the task of translating a text or speech in a source language to a different target language. Machine translation can be used to translate large volumes of text quickly without any human input. Machine translation datasets can be used to create MT models for different purpose such as: Internal emails and other written or oral communication. Documentation and instructions for products or services. Below is the list of African language datasets for Machine translation. 4. French to Ewe and French to Fongbe Machine Translation Dataset This is a parallel corpus dataset for machine translation from and . French to Ewe French to Fongbe Fonbge and Ewe are Niger-Congo languages, Fongbe is spoken in Benin with approximately 4.1 million speakers while Ewe is spoken in Togo and southeastern Ghana with approximately 4.5 million speakers. This dataset contains roughly French to Ewe and French to Fongbe parallel sentences, collected from blogs, tales, newspapers, daily conversations, webpages and annotated for neural machine translation. 23,000 53,000 5. Yorùbá to English Machine Translation Dataset This is a parallel sentence corpus dataset for machine translation from the Yorùbá language to the English language. Yorùbá is the Niger-Congo language and it is spoken in West Africa (southwestern Nigeria). The number of Yorùbá speakers is estimated at between 45 to 55 million. The dataset consists of parallel Yorùbá-English sentences from different domains like news, Yorùbá proverbs, movie transcript, localization translation, and books. 10,054 6. English to Luganda Machine Translation Dataset This is a parallel sentence corpus dataset for machine translation from the English language to the Luganda language. Luganda is a Bantu language and it is one of the major languages in Uganda. It is spoken by more than 8.5 million Baganda and other people in Kampala(capital city of Uganda). The dataset consists of parallel English-Luganda sentences and it was created by a team of researchers from the AI & Data Science research Lab at Makerere University with a team of Luganda teachers, students, and freelancers. 15,022 Sentiment Analysis Sentiment Analysis Datasets are used for the interpretation and classification of emotions ( ) within text data using different text analysis methods. positive, negative, and neutral Sentiment analysis has found its applications in various fields such as social media monitoring, brand monitoring, customer service, and market research. Below is the list of African language datasets for Sentiment Analysis. 7. Tunizi Dataset Tunizi is the first Tunisian Arabizi sentiment analysis dataset. Tunisian Arabizi represents the Tunisian dialect that is written in Latin characters and numbers rather than Arabic letters. gathered comments from social media platforms that express sentiment about popular topics. They extracted using public streaming APIs. iCompass 100k comments The collected comments were manually annotated using an overall polarity: Positive (1) Negative (-1) Neutral (0) The annotators were diverse in gender, age, and social background. You can also download this dataset from the : datasets python library datasets load_dataset

dataset = load_dataset( ) from import "tunizi" Speech Recognition Speech recognition, also known as Automatic Speech Recognition (ASR) can be defined as a technology that analyzes human speech and formulates an output, often a written transcription, in real-time. Sometimes referring to the process as "speech to text." Don't confuse this with voice recognition, as voice recognition just seeks to identify an individual user’s voice. Below is the list of African language datasets for Speech Recognition. 8. Speech Recognition dataset in Wolof Wolof is the language of Senegal, the Gambia, and Mauritania. It is spoken by more than 10 million people and about 40 percent (approximately 5 million people) of Senegal’s population speak Wolof as their native language. The ASR dataset has a total of audio files and transcriptions and it was created by a team of researchers from Baamtu Datamation company in Senegal. 6,683 9. Speech Recognition dataset in Kinyarwanda Kinyarwanda is the Bantu language and an official language of Rwanda. It is spoken by at least 12 million people in Rwanda, the Eastern Democratic Republic of the Congo, and southern Uganda. The was created by from different genders and ages in a common voice platform. The dataset has a total of of validated speech. The current dataset size is dataset 895 speakers 1,183 hours 40 GB. Topic Modeling Topic modeling uses unsupervised learning techniques to extract the main topic or set of topics that occur in a collection of text documents. Below is the list of African language datasets for Topic Modelling. 10. South African News Dataset This is the news dataset from South Africa. The news data were collected from SABC4 Facebook pages. The is the public broadcaster from South Africa. SABC The dataset contains news headlines (i.e short text) from and languages. Setswana is a Bantu language spoken in Southern Africa by about 8.2 million people while Sepedi is mainly spoken in the northern parts of South Africa by 4.7 million people. Setswana Sepedi Since the dataset is not annotated, you can use it to create a Topic model to cluster news data into different news topics such as sports, politics, culture, and entertainment. Final Thoughts on African Language Datasets I hope you found this list of different African language datasets useful and you can use them in your next . I will be happy to see what applications/solutions you will create from these datasets. If you couldn't find the dataset you need, please check out the following links: data science project Zenodo: African Natural Language Processing (AfricaNLP) Github: Masakhane Congratulations 👏👏, you have made it to the end of this article! I hope you have learned something new that will help you on your next data science project. If you learned something new or enjoyed reading this article, please share it so that others can see it. Until then, see you in the next post! You can also find me on Twitter . @Davis_McDavid And you can read more articles like this . here For more AI and machine learning guides, be sure to in the footer below. subscribe to our newsletter

Neural Labs Africa

neurallabs.africa

-- employees

Since n.d.

COMPANY RANKING

NEURAL LABS AFRICA

EVERGREEN INDEX #7550

Neural Labs Africa WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

Neural Labs Africa

neurallabs.africa

-- employees

Since n.d.

COMPANY RANKING

NEURAL LABS AFRICA

EVERGREEN INDEX #7550

RELATED COMPANIESRANK

STORIES ABOUT NEURAL LABS AFRICA

Neural Labs Africa WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!