Table 2 Multimodal classification performance using different input features.
ML classifier | Clinical features | Peak current changes | All 9 features | Selected 3 features | ||||
|---|---|---|---|---|---|---|---|---|
Accuracy (%) | AUC | Accuracy (%) | AUC | Accuracy (%) | AUC | Accuracy (%) | AUC | |
Decision tree | 57.01 | 0.56 | 81.30 | 0.84 | 89.72 | 0.85 | 90.65 | 0.86 |
(0.55–0.59) | (0.52–0.61) | (0.77–0.85) | (0.83–0.86) | (0.84–0.93) | (0.82–0.88) | (0.89–0.91) | (0.83–0.88) | |
k-nearest neighbors | 62.61 | 0.65 | 79.43 | 0.85 | 81.31 | 0.89 | 84.11 | 0.87 |
(0.59–0.66) | (0.63–0.69) | (0.77–0.82) | (0.83–0.86) | (0.77–0.84) | (0.88–0.89) | (0.83–0.88) | (0.85–0.90) | |
AdaBoost | 57.94 | 0.64 | 81.30 | 0.87 | 87.85 | 0.88 | 90.65 | 0.88 |
(0.57–0.60) | (0.61–0.67) | (0.79–0.83) | (0.86–0.88) | (0.85–0.89) | (0.87–0.90) | (0.89–0.91) | (0.84–0.92) | |
Naïve Bayes | 55.14 | 0.58 | 83.48 | 0.89 | 78.50 | 0.87 | 82.24 | 0.83 |
(0.52–0.55) | (0.56–0.60) | (0.79–0.87) | (0.85–0.93) | (0.75–0.81) | (0.83–0.91) | (0.79–0.86) | (0.81–0.86) | |
Random forest | 44.58 | 0.59 | 76.63 | 0.83 | 85.05 | 0.90 | 85.98 | 0.90 |
(0.42–0.46) | (0.55–0.63) | (0.74–0.79) | (0.81–0.85) | (0.80–0.90) | (0.87–0.92) | (0.80–0.89) | (0.87–0.93) | |
Neural network | 58.87 | 0.63 | 80.38 | 0.84 | 87.85 | 0.90 | 89.72 | 0.84 |
(0.56–0.60) | (0.62–0.64) | (0.76–0.84) | (0.81–0.87) | (0.83–0.90) | (0.87–0.92 | (0.84–0.93) | (0.82–0.85) | |
