Google AI can predict heart attack risk

Share

Google AI can predict heart attack risk

"The caveat to this is that it's early, (and) we trained this on a small data set", says Google's Lily Peng, a doctor and lead researcher on the project.

Scientists from Google and its affiliated research organization Verily have developed a new method of estimating risk of coronary heart disease with machine learning.

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. "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.

Sample retinal images highlighting important areas
Google AI can predict heart attack risk

It brings down the prediction time-frame to 5 years as against 10 years typically associated with the clinical risk predictors now in use. The traditional blood tests method is called the SCORE method and it can predict heart attacks with 72 percent accuracy, they report. "They're taking data that's been captured for one clinical reason and getting more out of it than we now do", said Oakden-Rayner. When given the retinal image of a patient who experienced a major cardiovascular event up to five years after the image was taken, and the image of a patient who did not, the algorithm could determine which patient experienced the health event 70% of the time. Discovering that we could do this is a good first step. Other disease factors such as diabetes significantly increases risk.

Verily trained these models using data from almost 300,000 patients, with the system then associating these factors together.

In order to analyse the working of the research, the scientists used machine learning on a medical dataset of nearly 300,000 patients, which included eye scans as well as general medical data. These techniques helped the company to generate a heatmap which basically shows which pixels were the most important for a predicting a specific CV risk factor. 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. This can give clinicians greater confidence in the algorithm, and potentially provide new insights into retinal features not previously associated with cardiovascular risk factors or future risk.

Share

Advertisement

© 2015 Leader Call. All Rights reserved.