Alphabet's Verily analyzing retinas w/ machine learning to predict heart disease

  • Alphabet's Verily analyzing retinas w/ machine learning to predict heart disease

Alphabet's Verily analyzing retinas w/ machine learning to predict heart disease

Google's parent company Alphabet has more than just the search giant under its banner; alongside its weird "X" experimental tech division sits its health science company Verily, which creates all manner of healthcare tech.

The new algorithms Google has introduced that involves human eye scanning are able to detect whether the person is suffering a condition of high blood pressure or at the risk of stroke or heart attack, according to the statement released by a team of Google researchers on Monday.

"Traditionally, medical discoveries are made by observing associations, making hypotheses from them and then designing and running experiments to test the hypotheses", reads the preamble to a paper published today on Nature Biomedical Engineering. "However, with medical images, observing and quantifying associations can be hard because of the wide variety of features, patterns, colors, values and shapes that are present in real images", researchers noted in a paper (PDF) published in the Nature journal Biomedical Engineering on Tuesday. A study which used data from 284,335 patients to train Google's deep learning algorithm, a variation of machine learning, to assess and predict cardiovascular disease risk based on changes in retinal images revealed a relatively high rate of accuracy.

It brings down the prediction time-frame to 5 years as against 10 years typically associated with the clinical risk predictors now in use. "We think that the accuracy of this prediction will go up a little bit more as we kind of get more comprehensive data".

Google also used some attention techniques to find out how the algorithm was making its prediction.

When presented with retinal images of two patients, one of whom suffered a cardiovascular event in the following five years, and one of whom did not, Google's algorithm was able to tell which was which 70 percent of the time.

Google researchers used scanned images of retinas of more than 280,000 patients to develop this algorithm. We found that each CV risk factor prediction uses a distinct pattern, such as blood vessels for blood pressure, and optic disc for other predictions. Results are most significant when the algorithm was tasked with determining specific risk factors. All of these factors are important predictors of cardiovascular health. For instance, the algorithm focuses on blood vessels when predicting blood pressure.

In the era of AI and machine learning, doctors are using patterns, generated by algorithms, to recognise diseases.