Table 4 Performance of one-class classification models in exercise detection for the COPD patients dataset using different predictor variables and performance indices
From: Machine-learning based feature selection for a non-invasive breathing change detection
Predictive variables | Accuracy | Sensitivity | Specificity | AUC |
|---|---|---|---|---|
Breathing rate | 0.594 | 0.537 | 0.615 | 0.592 (0.561-0.619) |
Signal amplitude | 0.678 | 0.577 | 0.715 | 0.685 (0.655-0.709) |
ARIMA coefficients | 0.634 | 0.610 | 0.642 | 0.654 (0.627-0.678) |
Breathing rate and signal amplitude | 0.650 | 0.629 | 0.658 | 0.683 (0.656-0.712) |
Breathing rate, signal amplitude and ARIMA coefficients | 0.644 | 0.629 | 0.649 | 0.686 (0.661-0.711) |
Fourier coefficients (frequencies ≤2 Hz) | 0.655 | 0.639 | 0.662 | 0.705 (0.681-0.731) |