Google AI Closer to Medical Breakthroughs

Google’s AI “Medical Brain” team has created an algorithm that can forecast outcomes ranging from how long people may stay in hospitals, their odds of re-admission and chances they might die soon. The neural network breakthrough could revolutionize the way doctors, hospitals and insurance companies make decisions in modern medicine.

A story in Bloomberg News yesterday outlined the stunning results of the company’s program and how it surpassed past efforts to mine and make use of data, largely due to tools that enabled the program to not only read PDFs, old chart and handwritten notes in medical files, but also to more accurately find and analyze data that humans would have trouble finding.

“More information shared and highlighted at the right time could save lives — and at the very least help medical workers spend less time on paperwork and more time on patient care,” according to the Bloomberg story. “But current methods of mining health data are costly, cumbersome and time consuming.”

In fact, 80% of work combing through files is simply attempting to make it usable. According to Google’s AI chief Jeff Dean, the algorithm will be able to predict symptoms and disease with a level of accuracy that some view as positive, other as frightening, since it may impact treatment. Dean said he thinks it will help doctors find the best diagnoses.

One result is a new trial in India that will use Google’s AI software to screen images of eyes for early signs of a condition called diabetic retinopathy.


Illustration by Maria Chimishkayan for Bloomberg Prognosis.