Loading the player...

Retinal Imaging, Algorithm Predicts Cardiovascular Disease Events

A deep learning algorithm that utilizes retinal images, Reti-CVD predicts cardiovascular disease (CVD) events in prediabetic and diabetic patients, according to research presented at the AHA Scientific Sessions 2023 in Philadelphia.

Tyler Hyungtaek Rim, MD, MBA, PhD, a vitreoretinal surgeon and chief medical officer at preventive healthcare startup Mediwhale in Singapore, spoke with MD /alert about the importance of the technology and the implications for cardiologists. Rim is also an adjunct assistant professor at SingHealth Duke NUS Academic Medical Center in Singapore.

The study included prediabetic and diabetic patients from the UK Biobank. The researchers assessed Reti-CVD score across three risk groups: low (n = 550), moderate (n = 276), and high (n = 275). Out of 1,101 prediabetic or diabetic patients at the onset, 12.5% (n = 138) experienced CVD events. Using Reti-CVD scores, the CVD events were categorized as low-risk in 8.2% (n = 45/550), moderate risk in 15.2% (n = 42/276), and high risk in 18.5% (n = 51/275) over a median follow-up period of 11 years.

The researchers found a significant association between the Reti-CVD score and the incidence of CVD events (hazard ratio [HR] = 1.57; 95% CI, 1-2.47 for the moderate-risk group; HR = 1.88; 95% CI, 1.19-2.98 for the high-risk group compared to the low-risk group).

Rim explains the importance of retinal imaging, the ease of implementation for Reti-CVD in primary care, and he gives a regulatory timeframe for the US.


Disclosures: Rim declared financial ties to Mediwhale. See full study for details.

Next Up