AI-Enabled ECG Algorithm Detects Heart Failure Early in Kenya
An AI-based electrocardiogram (ECG) algorithm has demonstrated strong performance in identifying early signs of heart failure among healthcare-seeking adults in Kenya. This finding comes from a recent multicenter study presented at Heart Failure 2025, addressing a critical need in Sub-Saharan Africa where heart failure affects younger patients and leads to worse outcomes compared to wealthier nations.
Why Early Detection Matters
Left ventricular systolic dysfunction (LVSD) is an early indicator of heart failure risk. Detecting LVSD allows for timely intervention, but the gold standard diagnostic tool—echocardiography—is often unavailable in resource-limited settings. The study explored whether an AI algorithm could assess LVSD using standard 12-lead ECGs, offering a cost-effective and scalable screening method.
Study Overview
- Conducted across eight healthcare facilities in Kenya
- Included 5,992 adult participants with a mean age of 55 years; 66% were female
- 65% classified as high cardiovascular risk based on prior cardiovascular disease or Framingham Risk Score (FRS) >10%
- All underwent 12-lead ECGs analyzed by the AI algorithm (AiTiALVSD; Medical AI Co, Seoul)
- Subset of 1,444 participants also had echocardiographic assessment for LVSD confirmation
Key Findings
The AI-ECG algorithm identified LVSD in 18.3% of participants overall. Prevalence was notably higher among those with high cardiovascular risk (22.9%) or established cardiovascular disease (32.0%) compared to those with low risk (9.9%). Echocardiography confirmed LVSD in 14.1% of the subset.
When compared to echocardiography, the AI algorithm showed:
- Sensitivity: 95.6%
- Specificity: 79.4%
- Negative Predictive Value: 99.1%
Implications for Healthcare Providers
This AI-ECG tool offers a practical alternative for screening LVSD in settings with limited access to echocardiography. It can identify at-risk individuals early, enabling timely intervention to potentially reduce progression to heart failure.
Given the high prevalence of undiagnosed LVSD detected, wider implementation of AI-ECG screening could have significant public health impact in Sub-Saharan Africa and similar regions.
Future research aims to expand screening across multiple African countries and to evaluate whether early identification through AI-ECG leads to increased use of proven heart failure treatments.
Learn More
For healthcare professionals interested in AI applications in cardiology and other fields, courses on AI technologies and their practical uses can be found at Complete AI Training.
More details on the study are available at the European Society of Cardiology’s website: Implementing an AI-ECG based algorithm to screen for left ventricular dysfunction in Kenya.
Your membership also unlocks:
