Indian scientists develop AI algorithm that can detect diabetes from ECG

The team from the Lata Medical Research Foundation in Nagpur included clinical data from 1,262 individuals. A standard 12-lead ECG heart trace lasting 10 seconds was performed for each participant. And 100 unique structural and functional characteristics for each lead were combined for each of the 10,461 single heartbeats recorded to generate a predictive algorithm called Diabeats.

Based on the size and shape of individual heartbeats, the DIABETES algorithm quickly detected diabetes and prediabetes with an overall accuracy of 97 percent and an accuracy of 97 percent, regardless of influential factors such as age, gender and co-existing metabolism. disorders.

Significant ECG features consistently correspond to known biological triggers that underlie the cardiac changes that are typical of diabetes and pre-diabetes.

If validated in larger studies, the approach could be used to investigate disease in low resource settings, the team said.

“Theoretically, our study provides a relatively inexpensive, non-invasive and accurate alternative (to current diagnostic methods) that can be used as a gatekeeper to effectively detect diabetes and pre-diabetes. May go.”

“Still, the adoption of this algorithm in routine practice will require robust validation on external, independent datasets,” he cautioned.

An estimated 463 million adults worldwide had diabetes in 2019. Detecting the disease at an early stage is important to prevent serious health problems later, but diagnosis relies heavily on blood glucose measurements.

In the paper published in the online journal BMJ Innovation, the researchers report that it is not only invasive, but also challenging to roll out as a large-scale screening test in low resource settings.

Structural and functional changes in the cardiovascular system are preceded by indicative blood glucose changes, and these are visible on the ECG heart marker.

The researchers also acknowledged that study participants were at high risk of diabetes and other metabolic disorders, so were unlikely to represent the general population. And diabetes was slightly less accurate among those taking prescription medications for diabetes, high blood pressure, high cholesterol, etc.

Nor were data available for people who became pre-diabetic or diabetic, making it impossible to determine the effectiveness of the initial screening.

Disclaimer: This story is auto-aggregated by a computer program and is not created or edited by FreshersLIVE.Publisher : IANS-media

Leave a Comment

Your email address will not be published.