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

Table 3 Performance evaluation of the model on the CFS dataset (n = 320) 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	BM-FE [19]	BM-DTS [19]	Sleep PPG-Net [19]
Accuracy	0.806	0.856	0.844	0.832	0.819	0.63	0.69	0.76
Cohen’s kappa (κ)	0.718	0.793	0.777	0.758	0.738	0.47	0.53	0.67
Weighted F1 score	0.8082	0.8574	0.8461	0.8337	0.8210	–	–	–

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

ISSN: 1472-6947