An artificial intelligence (AI) model can provide a “comprehensive characterisation” of diabetic cardiomyopathy, academics have said.
Research conducted at UT Southwestern Medical Center in Dallas has found that a machine learning model can detect individuals who are living with diabetic cardiomyopathy – a disorder of the heart muscle in people with diabetes which can lead to heart failure.
According to the results, AI can identify a high-risk phenotype of diabetic cardiomyopathy, potentially preventing heart failure as this would enable early interventions.
First author Dr Ambarish Pandey said: “This research is noteworthy because it uses machine learning to provide a comprehensive characterisation of diabetic cardiomyopathy – a condition that has lacked a consensus definition – and identifies a high-risk phenotype that could guide more targeted heart failure prevention strategies in people with diabetes.
- Patients prefer ChatGPT responses over doctors’ responses
- AI as effective as doctors for giving advice on back pain
- Nearly a quarter of GPs use AI to diagnose patients
“Phenotypes are observable physical properties of individuals that give them specific biological traits.”
As part of the trial, the scientists examined the health data of 1,000 adults from the Atherosclerosis Risk in Communities cohort, all of whom had diabetes but no history of cardiovascular disease.
The research team identified three patient subgroups by assessing a set of 25 echocardiographic parameters and cardiac biomarkers.
Approximately 27% of the participants made up the high-risk phenotype group, with these individuals exhibiting higher levels of NT-proBNP – a biomarker associated with abnormal heart remodelling and heart stress.
Participants in the high-risk phenotype group were 12.1% more likely to develop heart failure compared to those in the other subgroups.
Between 16% and 29% of people with diabetes have the high-risk phenotype, according to the findings.
Therefore, the research team has created a deep neural network classifier to detect more cases of diabetic cardiomyopathy.
- Blood test for children with obesity can flag up heart disease risk
- Low carbohydrate diets promote heart health according to new study
- High intake of saturated fat associated with heart damage
Dr Pandey said: “Clinically, this model could help target intensive preventive therapies, such as SGLT2 inhibitors, to patients most likely to benefit.
“It may also help enrich clinical trials of heart failure prevention strategies in people with diabetes.”
Dr Pandey added: “This builds on our previous work that evaluated the prevalence and prognostic implications of diabetic cardiomyopathy in community-dwelling adults.
“It extends those efforts by using machine learning to identify a more specific high-risk cardiomyopathy phenotype.”
Read the study in full in the European Journal of Heart Failure.