Table 3 Performance of one-class classification models in exercise detection for the healthy individuals 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 | 0.655 | 0.597 | 0.706 | 0.684 (0.647-0.721) |
Signal amplitude | 0.905 | 0.903 | 0.907 | 0.958 (0.942-0.971) |
ARIMA coefficients | 0.817 | 0.795 | 0.836 | 0.855 (0.828-0.880) |
Breathing rate and signal amplitude | 0.918 | 0.887 | 0.945 | 0.974 (0.964-0.981) |
Breathing rate, signal amplitude and ARIMA coefficients | 0.919 | 0.895 | 0.941 | 0.976 (0.967-0.983) |
Fourier coefficients (frequencies ≤2 Hz) | 0.929 | 0.951 | 0.909 | 0.971 (0.957-0.979) |