An enhanced hypergraph CNN with adaptive focal loss for automated ECG heartbeat classification
Akash Vijayan, Suchetha Manikandan, Deepak Joshi · 2026
Read An enhanced hypergraph CNN with adaptive focal loss for automated ECG heartbeat classification by Akash Vijayan, Suchetha Manikandan, Deepak Joshi (2026) 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-13444246.txt.
Description
Akash Vijayan and colleagues present a 2026 deep learning framework for automated heartbeat classification from ECG traces, tackling class imbalance, timing patterns, and beat-to-beat links that limit standard tools. Their hybrid pairs convolutional feature extraction with a hypergraph network built from nearest-neighbor beat relations, plus adaptive focal loss for rare classes.
The model scored 98.66 percent accuracy on the five-class MIT-BIH arrhythmia set and 95.46 percent on the three-class INCART set, holding steady across both benchmarks. That consistency suggests the beat-relation modeling captures patterns simpler models miss. Clinics gain a promising diagnostic aid, though prospective validation on hospital data remains.
Catalog record
| CREATOR | Akash Vijayan, Suchetha Manikandan, Deepak Joshi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |