US-based eye care solutions provider iHealthScreen has developed and validated its iPredict risk prediction models for cardiovascular diseases.

iPredict models are designed to offer artificial intelligence (AI)-based automated prediction scores for incident stroke and heart attack (myocardial infarction (MI)).

The data helps to identify individuals who are prone to getting stroke and MI within five years.

The iPredict models use an individual’s retinal colour images and health data to avert strokes and heart attacks, potentially saving lives from fatalities or disabilities.

In less than 60 seconds, the iPredict AI System is claimed to deliver fully automated reports on stroke and MI predictions. The entire testing process is completed in under five minutes, said iHealthScreen.

According to the eye care solutions provider, the prediction models showed improved overall accuracy for identifying individuals prone to stroke or heart attack within five years.

For incident stroke, iPredict demonstrated a sensitivity of 82.73%, a specificity of 87.35%, and an accuracy of 85.22%.

On the other hand, for incident MI, the models showed a sensitivity of 84.76%, a specificity of 84.67%, and an accuracy of 84.71%.

The company is planning to present these findings at the ESC Congress from 30 August to 02 September 2024.

Additionally, iPredict demonstrated superior accuracy compared to established stroke and heart attack prediction models like Framingham and CHADS/CHADS2VASC scores, iHealthScreen claimed.

Furthermore, these stroke and MI models by iPredict are designed to alert physicians in primary care settings about the necessity for further preventive measures among their patients.

iHealthScreen also offers AI-based screening tools for early detection of diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma.

The tools are CE-certified for the European market and approved by the UK’s Medicines and Healthcare products Regulatory Agency (MHRA), Australia’s Therapeutic Goods Administration (TGA), and UAE Health.

In May last year, the eye care solutions provider submitted 510K submission to the US Food and Drug Administration (FDA) for the iPredict tool for early diagnosis of AMD.