Follow for more talkers

Artificial intelligence accurately predicts diabetes

The tool named DiaBeats was almost 100 percent accurate

Avatar photo


Glucose meter with result of sugar level, blue circle of paper, fruits with tape measure, concept of slimming and symbol of world diabetes day
(Beyond This/Shutterstock)

By Mark Waghorn via SWNS

Diabetes could be predicted by artificial intelligence, according to new research.

An algorithm has been developed based on features of individual heartbeats recorded on an ECG (electrocardiogram).

The tool named DiaBeats was almost 100 percent accurate. It could lead to screening programs across the world.

The number of cases will reach half a billion within two decades. Picking up the disease in its early stages is key to preventing subsequent serious health problems.

Corresponding author Dr. Hemant Kulkarni, of Lata Medical Research Foundation,
Nagpur, India, said: "The non-invasive nature of ECG combined with the power of machine learning has potential for a screening method to detect type 2 diabetes and pre-diabetes."

But diagnosis relies heavily on the measurement of blood glucose. This is not only invasive but also challenging to roll out in primitive countries.

Structural and functional changes in the cardiovascular system occur early on even before indicative blood glucose changes - and these show up on an ECG heart trace.

The researchers enrolled participants in the Diabetes in Sindhi Families in Nagpur (DISFIN) study.

It looked at the genetic basis of type 2 diabetes and other metabolic traits. Participants provided details of their personal and family medical histories and their normal diet.

They underwent a full range of blood tests and clinical assessments. Their average age was 48 and 61 percent of them were women.

Pre-diabetes and diabetes were identified from the diagnostic criteria specified by the American Diabetes Association.

The prevalence of both type 2 diabetes and pre-diabetes was high - around 30 percent and 14 percent, respectively.

Insulin resistance was also high (35%)- as well as high blood pressure (51%), obesity (40%) and disordered blood fats (36%).

A standard 12-lead ECG heart trace lasting 10 seconds was done for each of the 1,262 volunteers.

And 100 unique structural and functional features for each lead were combined for each of the 10,461 single heartbeats recorded to generate a predictive algorithm.

It quickly detected diabetes and prediabetes with overall accuracy and precision of 97 percent - regardless of influential factors such as age, gender and coexisting metabolic disorders.

Important ECG features consistently matched the known typical biological triggers underpinning cardiac changes.

Dr. Kulkarni said: "In theory, our study provides a relatively inexpensive, non-invasive, and accurate alternative [to current diagnostic methods] which can be used as a gatekeeper to effectively detect diabetes and pre-diabetes early in its course.

“Nevertheless, adoption of this algorithm into routine practice will need robust validation on external, independent datasets."

More than 37 million Americans have diabetes (about 1 in 10), and approximately 90-95% of them have type 2 diabetes which is linked to unhealthy lifestyles.

Added Dr. Kukarni: "If externally validated, DiaBeats can be used as a gatekeeper to stratify individuals based on the risk of diabetes and pre-diabetes, especially in low-resource settings."

DiaBeats is described in BMJ Innovations.

Stories and infographics by ‘Talker Research’ are available to download & ready to use. Stories and videos by ‘Talker News’ are managed by SWNS. To license content for editorial or commercial use and to see the full scope of SWNS content, please email [email protected] or submit an inquiry via our contact form.

Top Talkers