Multi-Level Explainable AI for ECG-Based Atrial Fibrillation Detection: Exploring LIME, SHAP, and Grad-CAM for Clinical Interpretation
Jake Luo, Amirsajjad Taleban, Patrick Noffke, Rodney Sparapani · 2025
Read Multi-Level Explainable AI for ECG-Based Atrial Fibrillation Detection: Exploring LIME, SHAP, and Grad-CAM for Clinical Interpretation by Jake Luo, Amirsajjad Taleban, Patrick Noffke, Rodney Sparapani (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13112545.txt.
Description
Jake Luo, Amirsajjad Taleban, Patrick Noffke, and Rodney Sparapani built an AI system that flags a common heart rhythm fault from ECG traces. The system also explains its own decisions in ways doctors can use. Deep learning reads ECGs well, but its black-box nature blocks clinical trust. Explanations must match how clinicians actually reason about heart traces.
The model was trained on nearly six thousand ECG samples from a public data set. It caught the rhythm fault with high precision and recall, and sorted normal rhythm even better. Three explanation tools highlighted what drove each call. The team found the usual tools too fine-grained for clinical use and built coarser, layered views. Their framework aims at AI that a cardiologist can audit.
Catalog record
| CREATOR | Jake Luo, Amirsajjad Taleban, Patrick Noffke, Rodney Sparapani |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |