This Is How AI Can Help Detect Cancer

The U.S. saw close to 5,000 more cancer deaths in 2023 compared with 2022, marking the only cause of death among the top 10 to see an increase year over year, according to a U.S. News and World report. After heart disease, which comes in as the No. 1 killer of Americans, cancer remained the No. 2 killer in 2023, accounting for 19.8% of all deaths.

What this means is that we need to find solutions to help prevent and detect these top causes of death in America. And this is where AI comes in.

With AI, doctors can create 3D models from 2D images, providing a more comprehensive view of tumors and their potential spread, improving the accuracy of diagnosis and treatment planning.

AI can be used to assist in analyzing tissue samples by examining microscopic images of biopsy samples. It can recognize cancerous cells and differentiate between different types of cancer with a high degree of accuracy.

By using advanced AI tools, doctors will be able to automate the counting and classification of cells in tissue samples, speeding up the diagnostic process and reducing the workload on pathologists.

Where AI can truly shine is through predictive analytics. AI models can be trained on large datasets to predict an individual’s risk of developing certain types of cancer-based on their genetic, lifestyle, and environmental factors.

Using these large datasets as a basis, AI can then analyze a patient’s clinical data to predict the likely course of the disease and the patient’s response to treatment, helping clinicians tailor treatment plans more effectively.

Naturally, AI will reduce errors and bias by reducing diagnostic errors caused by human fatigue or oversight. This leads to more accurate and consistent cancer detection.

When it comes to research and development, AI can accelerate the discovery of new cancer treatments by analyzing vast amounts of biological and chemical data, identifying potential drug candidates faster than traditional methods.

This isn’t just speculation.

AI is “outpacing doctors when it comes to detecting a common cancer in men,” according to a recent report.

A new study from UCLA found that “AI identified prostate cancer with 84% accuracy — compared to 67% accuracy for cases detected by physicians.”

The AI software program, Unfold AI, made by Avenda Health in California, “uses an AI algorithm to visualize the likelihood of cancer based on various types of clinical data.”

In further news, an AI model has identified certain breast tumor stages likely to progress to invasive cancer.

According to a report in MIT News, “Ductal carcinoma in situ (DCIS) is a type of preinvasive tumor that sometimes progresses to a highly deadly form of breast cancer. It accounts for about 25 percent of all breast cancer diagnoses.

“Because it is difficult for clinicians to determine the type and stage of DCIS, patients with DCIS are often overtreated. To address this, an interdisciplinary team of researchers from MIT and ETH Zurich developed an AI model that can identify the different stages of DCIS from a cheap and easy-to-obtain breast tissue image.”

One AI model even helps researchers detect disease based on coughs.

Google has developed a foundation model called Health Acoustic Representations (HeAR), which is trained on a massive dataset of health-related sounds. The technology described involves using bio-acoustic sounds—such as coughs, speech, and breathing—as indicators of health conditions, analyzed through AI.

HeAR is designed to recognize patterns in these sounds, which can be used to detect diseases like tuberculosis (TB) and chronic obstructive pulmonary disease (COPD).

The key advantage of HeAR is its ability to accurately analyze health-related sounds using less training data, making it highly efficient in the data-scarce field of healthcare. This model can be generalized across different microphones, enhancing its usability in various settings.

HeAR's strength lies in its ability to accelerate the development of custom bio-acoustic models for specific health conditions, even in cases where data is limited, or resources are scarce.

An example of its application is seen in the work of Salcit Technologies, which uses HeAR to improve its AI-based tool, Swaasa®, for the early detection of TB through cough analysis. This approach could significantly enhance the accessibility and affordability of TB screening, particularly in regions like India, where TB remains a major public health challenge.

By leveraging AI to analyze sounds produced by the human body, HeAR has the potential to revolutionize disease screening and monitoring, offering a low-cost, scalable solution that can be deployed widely, even in resource-limited settings. This technology could play a crucial role in global health initiatives, such as the effort to eradicate TB by 2030, by making early diagnosis more accessible and improving health outcomes on a large scale.

The future is already here with AI being used on numerous applications to predict or detect cancer. Further technological developments will only assist us in preventing more deaths and offering more Americans the quality of life they might not have had until now.

All thanks to AI.