2-class respiratory sound classification on ICBHI 60-40% split official (test)
81.66SpecificityBae et al. (Patch-Mix CL)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Bae et al. (Patch-Mix CL)architecture=AST, pretrain=IN + AS2023.12 | 81.66 | 55.77 | 68.71 | |
| Patch-Mix CL (ours)architecture=AST, pretrain=IN + AS2023.05 | 81.66 | 55.77 | 68.71 | |
| Bae et al. (Patch-Mix CL)Backbone=AST, Pretraining=IN + AS, Metadata=X2025.05 | 81.66 | 55.77 | 68.71 | |
| Bae et al. (Patch-Mix CL)Backbone=AST, Pretraining Data=IN + AS, Venue=INTERSPEECH‘232026.06 | 81.66 | 55.77 | 68.71 | |
| SG-SCLarchitecture=AST, pretrain=IN + AS2023.12 | 79.87 | 57.97 | 68.93 | |
| Kim et al. (SG-SCL)Backbone=AST, Pretraining=IN + AS, Metadata=O2025.05 | 79.87 | 57.97 | 68.93 | |
| Kim et al. (SG-SCL)Backbone=AST, Pretraining Data=IN + AS, Venue=ICASSP‘242026.06 | 79.87 | 57.97 | 68.93 | |
| DASS (Lung-SRAD)Backbone=DASS, Pretraining Data=AS, Venue=-2026.06 | 79.53 | 65.61 | 72.57 | |
| Nguyen et al. (CoTuning)architecture=ResNet50, pretrain=IN2023.12 | 79.34 | 50.14 | 64.74 | |
| Nguyen et al. (CoTuning)architecture=ResNet50, pretrain=IN2023.05 | 79.34 | 50.14 | 64.74 | |
| Nguyen et al. (CoTuning)Backbone=ResNet50, Pretraining=IN, Metadata=X2025.05 | 79.34 | 50.14 | 64.74 | |
| Nguyen et al. (StochNorm)architecture=ResNet50, pretrain=IN2023.05 | 78.86 | 49.79 | 64.32 | |
| Bae et al. (Fine-tuning)architecture=AST, pretrain=IN + AS2023.12 | 77.14 | 56.4 | 66.77 | |
| Fine-tuning (ours)architecture=AST, pretrain=IN + AS2023.05 | 77.14 | 56.4 | 66.77 | |
| Bae et al. (Fine-tuning)Backbone=AST, Pretraining Data=IN + AS, Venue=INTERSPEECH‘232026.06 | 77.14 | 56.4 | 66.77 | |
| DATarchitecture=AST, pretrain=IN + AS2023.12 | 77.11 | 56.98 | 67.04 | |
| DASS (Spectral-Aware Regularization)Backbone=DASS, Pretraining Data=AS, Venue=-2026.06 | 76.72 | 60.29 | 68.4 | |
| BEATS + CEBackbone=BEATS, Pretraining=AS, Metadata=X2025.05 | 75.19 | 66.34 | 70.76 | |
| BEATS + PAFABackbone=BEATS, Pretraining=AS, Metadata=X2025.05 | 74.87 | 68.29 | 72.08 | |
| Jeong et al. (PAFA)Backbone=BEATs, Pretraining Data=AS, Venue=INTERSPEECH‘252026.06 | 74.87 | 68.29 | 72.08 | |
| DASS (Fine-tuning)Backbone=DASS, Pretraining Data=AS, Venue=-2026.06 | 73.74 | 61.72 | 68.2 | |
| CNN-MoEarchitecture=C-DNN2023.12 | 72.4 | 37.5 | 54.1 | |
| CNN-MOEarchitecture=C-DNN2023.05 | 72.4 | 37.5 | 54.1 |