InsightSleepNet: the interpretable and uncertainty-aware deep learning network for sleep staging using continuous Photoplethysmography

Table 4 Performance evaluation of the model on the CAP dataset (n = 24) and a comparison before and after energy score thresholding

	InsightSleepNet (without thresholding)	0.80 Energy threshold	0.85 Energy threshold	0.90 Energy threshold	0.95 Energy threshold	Zhao et al., 2021 [44]
Accuracy	0.806	0.843	0.835	0.825	0.843	0.77
Cohen’s kappa (κ)	0.730	0.786	0.775	0.760	0.786	0.69
Weighted F1 score	0.8046	0.8417	0.8342	0.8242	0.8133	–

The ‘-’ symbol indicates that the corresponding metric is not provided in the study

ISSN: 1472-6947