From the retinal images, Google's AI can determine within impressive degrees of accuracy a patient's age, gender, blood pressure, and smoking status, as well as even the past occurrence of major cardiovascular events, The Verge explains. Once trained, they tested the method by having it look at two eye scans: one normal, and one from a patient who suffered a cardiac event within five years of the image being taken.
Traditionally, medical discoveries are often made through a sophisticated form of guess and test making hypotheses from observations and then designing and running experiments to test the hypotheses. The latest applies machine learning to retinal images to identify the risk factors of cardiovascular disease.
In the era of AI and machine learning, doctors are using patterns, generated by algorithms, to recognise diseases. It was also able to deduce a persons age, blood pressure, and whether they smoke.
The study was published in Nature Biomedical Engineering, and Google Brain Team Product Manager Lily Peng, one of the researchers behind the study, also wrote up a blog entry on the study for Google Research. Michael V McConnell, head of Cardiovascular Health Innovations at Verily, said that the research needs more work and a larger patient database to validate these findings before it's ready for clinical testing.
"Cardiovascular disease is the leading cause of death globally". "We think that the accuracy of this prediction will go up a little bit more as we kind of get more comprehensive data". Using this data, the company's software can predict the risk of cardiovascular diseases such as heart attack or stroke.
"They're taking data that's been captured for one clinical reason and getting more out of it than we now do", Luke Oakden-Rayner, a medical researcher from the University of Adelaide, told The Verge.
Google AI's method reportedly uses deep learning algorithms to create a so-called "heat map or graphical representation of data which revealed which pixels in an image". All of these factors are important predictors of cardiovascular health. So, for example, if most patients that have high blood pressure have more enlarged retinal vessels, the pattern will be learned and then applied when presented just the retinal shot of a prospective patient.