Predicting Cardiovascular Risk Factors from Retinal Fundus Photographs using Deep Learning

TL;DR

This study employs deep learning on retinal fundus images to predict cardiovascular risk factors and events, revealing new associations and anatomical insights, with high accuracy validated across multiple datasets.

Contribution

It introduces a deep learning approach that predicts cardiovascular risk factors from retinal images, uncovering new, quantifiable associations previously difficult to observe.

Findings

Accurately predicts age, gender, smoking status, HbA1c, systolic blood pressure, and cardiac events.

Models utilize distinct retinal features like optic disc and blood vessels.

Validated on large, independent datasets with high performance metrics.

Abstract

Traditionally, medical discoveries are made by observing associations and then designing experiments to test these hypotheses. However, observing and quantifying associations in images can be difficult because of the wide variety of features, patterns, colors, values, shapes in real data. In this paper, we use deep learning, a machine learning technique that learns its own features, to discover new knowledge from retinal fundus images. Using models trained on data from 284,335 patients, and validated on two independent datasets of 12,026 and 999 patients, we predict cardiovascular risk factors not previously thought to be present or quantifiable in retinal images, such as such as age (within 3.26 years), gender (0.97 AUC), smoking status (0.71 AUC), HbA1c (within 1.39%), systolic blood pressure (within 11.23mmHg) as well as major adverse cardiac events (0.70 AUC). We further show that our models used distinct aspects of the anatomy to generate each prediction, such as the optic disc or blood vessels, opening avenues of further research.