Machine-learning based feature selection for a non-invasive breathing change detection

Table 1 Performance of supervised classification models in exercise detection for the healthy individuals dataset using different predictor variables and performance indices

Predictive variables	Accuracy	Sensitivity	Specificity	AUC
Breathing rate	0.886	0.993	0.282	0.734 (0.673-0.794)
Signal amplitude	0.957	0.986	0.795	0.987 (0.978-0.995)
ARIMA coefficients	0.859	0.959	0.295	0.820 (0.769-0.872)
Breathing rate and signal amplitude	0.965	0.984	0.859	0.995 (0.991-1.000)
Breathing rate, signal amplitude and ARIMA coefficients	0.963	0.979	0.872	0.977 (0.945-1.000)
Fourier coefficients (frequencies ≤2 Hz)	0.954	0.973	0.846	0.975 (0.948-1.000)

ISSN: 1756-0381