In a paradigm shift for the health tech industry, Sundar Pichai, CEO of Google, has revealed a groundbreaking development in artificial intelligence (AI) technology. Google’s AI can now predict cardiovascular events by analyzing eye scans, potentially offering an alternative to conventional diagnostic methods such as CT scans, MRIs, and X-rays.
The journey began four years ago when a team of researchers from Google and Aravind Eye Hospital collaborated to develop an automated tool for detecting diabetic retinopathy, a leading cause of blindness worldwide. This initial algorithm, capable of recognizing signs of the disease, has now advanced to deliver diagnoses within seconds when provided with retinal photos. Soon, it is expected to operate independently, transforming the landscape of eye disease detection and management.
Expanding on its capabilities, Google recently introduced an algorithm that can identify a person’s sex, smoking status, and predict the five-year risk of a heart attack based solely on retinal imagery. The AI’s ability to detect problems beyond human perception opens possibilities for early detection of various diseases, including dementia, multiple sclerosis, Parkinson’s, Alzheimer’s, and even schizophrenia.
Eye, particularly the retina, serves as a window into overall health
The eye, particularly the retina, serves as a window into overall health. By examining the appearance of blood vessels in the retina, doctors can infer critical information like blood pressure, age, and smoking habits, which are crucial predictors of cardiovascular health. To train their cardiovascular prediction algorithm, Google and Verily’s scientists employed machine learning to analyze a dataset of almost 300,000 patients, incorporating both eye scans and general medical data.
When tested, Google’s AI successfully distinguished between retinal images of patients who experienced cardiovascular events within five years and those who did not, with an accuracy rate of 70%. This performance closely matched the commonly used SCORE method, which requires a blood test and achieves correct predictions 72% of the time.
This groundbreaking method of assessing cardiovascular risk not only simplifies and accelerates diagnostics for doctors but also paves the way for an AI-powered era of scientific discovery. With enough data, artificial intelligence could generate entirely new medical insights autonomously. Moreover, this technology holds promise for remote areas without access to specialized medical equipment. A smartphone, inexpensive condensing lens, and DIY retinal camera could enable vision screenings on the go, with patients receiving rapid diagnoses through cloud-based image analysis.
As Sundar Pichai’s announcement highlights, Google’s AI is already advancing toward a future where traditional diagnostic methods may become secondary to the insights gleaned from eye scans. A quick eye scan could potentially uncover hidden health issues and predict cardiovascular events, transforming healthcare and diagnostics. The vision of healthcare is becoming clearer as Google’s AI deciphers the stories told by our eyes.