Artificial intelligence and heart traces can predict diabetes and pre-diabetes in patients, preliminary research suggests.
The combination of an algorithm made for individual heartbeats and electrocardiogram (ECG) tracking accurately predicted diabetes patients, a study published in the online journal BMJ Innovations found.
If validated by larger studies, the approach could be used to screen for the disease in low resource settings, the researchers 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,” they said.
“Nevertheless, adoption of this algorithm into routine practice will need robust validation on external, independent data sets.”
About 463 million adults worldwide had diabetes in 2019, and diagnosis relies heavily on the invasive procedure of measuring blood glucose. That is also difficult in low resource regions of the world.
Detecting the disease in its early stages is vital in preventing subsequent serious health problems.
Researchers wanted to see if AI techniques could be used to harness the screening potential of ECG to predict pre-diabetes and type 2 diabetes in people at high risk of the disease.
They looked at participants in the Diabetes in Sindhi Families in Nagpur (DISFIN) study, which monitored families at high risk of type 2 diabetes in Nagpur, India.
The prevalence of both type 2 diabetes and pre-diabetes was high: about 30 per cent and 14 per cent, respectively.
The prevalence of insulin resistance was also high at 35 per cent — as was the case with other influential coexisting conditions, including high blood pressure, obesity, and disordered blood fats.
A standard 12-lead ECG heart trace lasting 10 seconds was carried out for each of the 1,262 participants. And for each lead, 100 unique structural and functional features were combined for each of the 10,461 single heartbeats recorded to generate the predictive algorithm.
Based on the shape and size of individual heartbeats, the DiaBeats algorithm quickly detected diabetes and pre-diabetes with an overall accuracy of 97 per cent and a precision of 97 per cent. This was irrespective of influential factors, such as age, gender, and coexisting metabolic disorders.
ECG features consistently matched the known biological triggers underpinning cardiac changes that are typical of diabetes and pre-diabetes.
The team acknowledged that the study participants were all at high risk of diabetes and other metabolic disorders, so unlikely to represent the general population.
