Table 7 Comparative study of different algorithms.
From: Deep Learning-Driven Early Diagnosis of Respiratory Diseases using CNN-RNN Fusion on Lung Sound Data
Model | Dataset | Lung Disease | Precision | Recall | F1 Score | Accuracy |
|---|---|---|---|---|---|---|
Deep Learning (Proposed) | ICBHI | Healthy | 0.9713 | 0.9827 | 0.9811 | 0.9726 |
Pneumonia | 0.9472 | 0.9318 | 0.9335 | 0.9356 | ||
Asthma | 0.9219 | 0.9548 | 0.9322 | 0.9445 | ||
COPD | 0.9115 | 0.9022 | 0.9047 | 0.9064 | ||
Overall | 0.9441 | 0.943 | 0.9426 | 0.9435 | ||
Coswara | Healthy | 0.9152 | 0.9711 | 0.9437 | 0.9523 | |
COVID-19 | 0.8927 | 0.8239 | 0.8541 | 0.8634 | ||
Overall | 0.9026 | 0.9228 | 0.9117 | 0.9189 | ||
Support Vector Machine | ICBHI | Healthy | 0.8623 | 0.9702 | 0.9132 | 0.9247 |
Pneumonia | 0.8714 | 0.7916 | 0.8225 | 0.8269 | ||
Asthma | 0.8122 | 0.8743 | 0.8416 | 0.8421 | ||
COPD | 0.7805 | 0.7811 | 0.7813 | 0.7824 | ||
Overall | 0.8373 | 0.8645 | 0.8571 | 0.8614 | ||
Random Forest | Coswara | Healthy | 0.8245 | 0.9628 | 0.8813 | 0.8939 |
COVID-19 | 0.7733 | 0.6278 | 0.6931 | 0.7214 | ||
Overall | 0.8022 | 0.879 | 0.8415 | 0.8546 |
