BELGRADE, Serbia — A new study from Kenya shows that an artificial intelligence algorithm analyzing electrocardiograms (AI-ECG) can effectively detect subclinical heart failure in patients, even when advanced imaging like echocardiography is not readily accessible.
At the European Society of Cardiology’s Heart Failure 2025 congress, Dr. Ambarish Pandey of UT Southwestern Medical Center presented findings demonstrating that the AI-ECG model accurately identified left ventricular systolic dysfunction—defined as an ejection fraction below 40%. The algorithm showed strong sensitivity, specificity, and a high negative predictive value.
Patients flagged as positive by the AI-ECG also displayed signs of adverse cardiac changes, such as left ventricular hypertrophy and diastolic dysfunction, which are indicators of worsening heart function.
“These results highlight the potential for AI-ECG-based screening tools to identify left ventricular systolic dysfunction in areas where resources are limited,” Dr. Pandey said.
Independent expert Dr. Christian Mueller from the Cardiovascular Research Institute Basel noted that the technology could benefit wealthier nations as well. “Detecting this form of heart failure remains a challenge worldwide. AI-ECG could fill an important gap even in resource-rich countries,” he commented.
Related topics:
