Table 4 Performance of different machine learning algorithms in the classification of moderate COPD and severe COPD
From: Multi-channel lung sounds intelligent diagnosis of chronic obstructive pulmonary disease
Model | Accuracy | Sensitivity | Specificity | AUC | F1-Score | Kappa |
|---|---|---|---|---|---|---|
SVM | 94.26 | 97.32 | 89.93 | 97.54 | 94.25 | 88.16 |
(92.70–95.85) | (96.83–98.01) | (87.79–92.16) | (96.96–98.62) | (93.16–95.03) | (85.85–90.42) | |
Bayes | 89.37 | 99.17 | 79.61 | 97.75 | 90.30 | 78.74 |
(87.70–91.04) | (98.74–99.61) | (76.48–82.73) | (97.06–98.44) | (88.68–91.92) | (75.52–81.96) | |
Decision Tree | 70.63 | 72.69 | 68.56 | 68.04 | 71.20 | 41.24 |
(69.07–72.18) | (70.13–75.25) | (65.09–72.04) | (66.56–69.52) | (70.17–72.22) | (37.79–44.69) | |
DBN | 83.75 | 87.80 | 79.66 | 85.01 | 84.42 | 67.45 |
(80.71–86.78) | (83.72–91.87) | (74.25–85.06) | (82.01–88.01) | (81.27–87.57) | (61.40–73.49) |