Utilizing Machine Learning to Classify Advanced Rheumatic Heart Disease via Electrocardiogram in Cardiology Settings
-
By
-
Amsalu Tomas Chuma
-
Melkamu Hunegnaw Asmare
-
Carolina Varon
-
Desalew Mekonnen Kassie
-
Chunzhuo Wang
-
Bart Vanrumste
-
December 1, 2025
-
Clinical Scorecard: Utilizing Machine Learning to Classify Advanced Rheumatic Heart Disease via Electrocardiogram in Cardiology Settings
At a Glance
| Category | Detail |
|---|
| Condition | |
| Key Mechanisms | Progression from untreated group A streptococcal infections to acute rheumatic fever, leading to heart valve damage and autoimmune response. |
| Target Population | |
| Care Setting | |
Key Highlights
- RHD is a significant cause of morbidity and mortality, particularly in Sub-Saharan Africa.
- Echocardiography is the gold standard for RHD diagnosis, but ECG remains underutilized.
- Machine learning models can enhance RHD detection using ECG signals in resource-limited settings.
- Early intervention with benzathine penicillin G (BPG) injections is crucial.
Guideline-Based Recommendations
Diagnosis
- Echocardiographic criteria are recommended for RHD diagnosis.
- Consider integrating ECG as a supplementary diagnostic tool.
Management
- Early intervention with benzathine penicillin G (BPG) injections is crucial.
- Consider additional management strategies based on patient history.
Monitoring & Follow-up
- Regular follow-up and monitoring of patients with a history of rheumatic fever.
- Implement structured monitoring protocols.
Risks
- Inadequate healthcare resources and low awareness of ARF increase RHD burden.
Patient & Prescribing Data
Young adults and children in high-risk areas, defined by socioeconomic and health access factors.
BPG injections can significantly reduce disease progression; consider regular follow-ups.
Clinical Best Practices
- Implement primary prevention and screening strategies in high-risk populations.
- Utilize machine learning for improved diagnostic accuracy in RHD detection.
References
