Respiratory sound 4-class classification on ICBHI 2017 (80/20 random split)
84.1SpecificityAcharya et al.
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Acharya et al.2020.10 | 84.1 | 48.6 | 66.3 | |
| RespireNet (CNN+CBA+BRC+FT)architecture=CNN, augmentation=Concatenation-based augmentation, clipping=Blank region clipping, fine-tuning=device specific2020.10 | 83.3 | 53.7 | 68.5 | |
| RespireNet (CNN+CBA+BRC)architecture=CNN, augmentation=Concatenation-based augmentation, clipping=Blank region clipping2020.10 | 79.7 | 54.4 | 67.1 | |
| RespireNet (CNN base)architecture=CNN2020.10 | 78.8 | 53.6 | 66.2 | |
| Kochetov et al.2020.10 | 73 | 58.4 | 65.7 | |
| Ma et al. [12]2020.10 | 64.7 | 63.7 | 64.2 |