Table 3 Classification Performance on ICBHI Dataset.
From: Deep Learning-Driven Early Diagnosis of Respiratory Diseases using CNN-RNN Fusion on Lung Sound Data
Disease | Accuracy (%) | Precision (%) | Recall (%) | F1-Score (%) | AUC |
|---|---|---|---|---|---|
Normal | 95.2300 ± 1.2400 | 94.7800 ± 1.1100 | 95.6600 ± 1.4500 | 95.2200 ± 1.0800 | 0.9620 ± 0.0120 |
COPD | 93.8900 ± 1.5600 | 92.3100 ± 1.3400 | 94.2100 ± 1.6500 | 93.2500 ± 1.4400 | 0.9510 ± 0.0140 |
Bronchiectasis | 91.6700 ± 1.3800 | 90.8400 ± 1.1200 | 91.0200 ± 1.4500 | 90.9200 ± 1.3100 | 0.9460 ± 0.0130 |
Pneumonia | 92.4100 ± 1.4400 | 93.1100 ± 1.2300 | 91.5500 ± 1.6500 | 92.3200 ± 1.1800 | 0.9480 ± 0.0110 |
Asthma | 90.8900 ± 1.5500 | 89.7700 ± 1.4500 | 91.2000 ± 1.5200 | 90.4700 ± 1.4900 | 0.9390 ± 0.0150 |
Average | 92.8200 ± 1.4300 | 92.1600 ± 1.2500 | 92.7300 ± 1.5400 | 92.4300 ± 1.3000 | 0.9490 ± 0.0130 |
