Dive Brief:
- Researchers at Alphabet and its research arm Verily Life Sciences have found a way to predict a person’s cardiovascular risk factors using eye scans and deep learning, according to a study published this week in Nature Biomedical Engineering.
- By analyzing scans of the retinal fundus, the tissue at the back of a patient’s eye, the company’s software can tease out data such as blood pressure, age and whether an individual smokes — all potential risk factors for having a major cardiac event.
- The algorithm could speed assessments of patients’ cardiovascular risk, but more testing is required before it can be used in a clinical setting.
Dive Insight:
“Using deep learning algorithms trained on data from 284,335 patients, we were able to predict CV risk factors from retinal images with surprisingly high accuracy for patients from two independent data sets of 12,026 and 999 patients,” Dr. Lily Peng, product manager of the Google Brain Team and a co-author of the study, wrote in the Google Research Blog.
“For example, our algorithm could distinguish the retinal images of a smoker from that of a non-smoker 71% of the time, compared to a ~50% (i.e. random) accuracy by human experts,” she said.
The results also show “strong gender differences” in the fundus images that could help guide research on the differences in male and female eyes, as well as how cardiovascular disease or risk factors affect retinal health, according to the study.
Last year, Verily began recruiting 10,000 volunteers to help build a comprehensive database of biometric data. The company is also working with French drugmaker Sanofi to develop tools for diabetes management and with 3M on solutions for population health.
This latest study could have implications for treating patients. Cardiology has been trending toward population health and tools such as this new algorithm could help identify and inform care paths for patients.