Apollo Radiology deploying AI models to improve health outcomes in India

One of our biggest opportunities with AI in healthcare is the ability to increase access to high-quality, equitable care across the globe. Radiology is one area where expertise is critical to patient outcomes, but there aren’t always enough radiologists to meet patients’ needs. This is a situation where AI can make a difference.

Our newest collaboration with Apollo Radiology International in India is building on these AI advancements in healthcare and bringing them to communities across the country.

AI supported early detection for high-mortality diseases
More than 10 million people a year become ill with tuberculosis (TB), and more than 1.3 million die from the disease worldwide annually, with particularly high rates in South Asia and Sub-Saharan Africa. TB can be treated and cured, but delays in treatment allow the disease to spread through the community and develop to the point of being fatal in patients.

Chest X-rays are a common method of screening for the disease, but there are not enough trained radiologists to interpret images at scale. Further, TB is usually diagnosed with tests that are often not available in rural settings or areas where there are few hospitals and healthcare centers. To help, we’ve developed an AI system to interpret chest X-ray scans for early signs of TB.

Lung cancer and breast cancer also require specialized screening and medical expertise to interpret scans. In India, lung cancer is one of the leading causes of cancer death. There is no nationalized screening program for high-risk individuals to try to catch the disease early, when treatment is most effective. Lung cancer can also be identified incidentally if a patient happens to get a scan for some other reason. However, radiologists need to be trained to identify these incidental nodules and have time to look for them.

AI can help on both fronts, making screening more broadly accessible and providing an additional check to identify incidental nodules and follow-up as needed.

India also has more than three times the death rate from breast cancer, compared to the U.S. There is a shortage of radiologists trained to identify breast cancer and screening is not widely accessible. We’re hopeful that AI can assist radiologists in interpreting mammograms to scale screening in India.

What’s next for AI-powered screening
AI can play a tremendous role in the early detection of all these diseases. We’ve been working with Apollo Radiology International over the past few years to validate our AI systems in regions where they can have the most impact. Today, we continue to collaborate with Apollo as it works towards securing regulatory approvals to bring these models into clinical care in communities across India.

Over the next 10 years, Apollo Radiology International will use these models to provide 3 million free AI-powered screenings for TB, lung cancer, and breast cancer — helping hundreds of thousands more people across India access timely care.

How we support efforts to improve maternal and child health
Our work in India also extends to helping others develop solutions. With the support of Google.org, ARMMAN is tackling maternal and child mortality and morbidity with the creation of mMitra. This is a free mobile service that sends timely and targeted preventive care messages to new and expectant mothers. Using Google’s AI powered predictions health workers are able to follow-up with women at risk of dropping out of the health program. Google Blog