Classify murmurs 벤치마크
Classify murmurs on CirCor DigiScope
Weighted Accuracy
- 2022-09-07 — Inception Time: Weighted Accuracy 0.593
- 2023-05-26 — DBResNet: Weighted Accuracy 0.771
- 2024-04-26 — M2D: Weighted Accuracy 0.832
| Rank | Model | Weighted Accuracy | Weighted accuracy (validation) | Weighted accuracy (cross-val) | Unweighted average recall | Paper | Code | Year |
|---|---|---|---|---|---|---|---|---|
| 1 | M2D | 0.832 | – | – | 0.713 | Exploring Pre-trained General-purpose Audio Representations for Heart Murmur Detection | nttcslab/m2d · nttcslab/eval-audio-repr | 2024 |
| 2 | DBResNet | 0.771 | 0.768 | – | – | Dual Bayesian ResNet: A Deep Learning Approach to Heart Murmur Detection | Benjamin-Walker/heart-murmur-detection | 2023 |
| 3 | Inception Time | 0.593 | 0.522 | 0.497±0.083 | – | Phonocardiogram Classification Using 1-Dimensional Inception Time Convolutional Neural Networks | Bsingstad/Heart-murmur-detection-2022-Simulab | 2022 |