A Novel Approach for Sleep Arousal Disorder Detection Based on the Interaction of Physiological Signals and Metaheuristic Learning

Table 6

The performance of models for arousal detection and sleep stage classification using metrics such as accuracy and computational complexity.

Author	Procedure type	Signal (s)	Accuracy (%)	Computational complexity
Supratak et al. [56]	Sleep stage scoring	Raw single-channel EEG	85.50	High
Willemen et al. [57]	Sleep stage classification	Cardiorespiratory and movement	86.75	Moderate
Fonseca et al. [58]	Sleep stage detection	Cardiorespiratory	87.38	High
Alickovic et al. [59]	Sleep stage classification	EEG	91.10	Moderate
Fernández-Varela et al. [60]	Arousals monitoring	Polysomnographic	86.00	Low
Li and Guan [25]	Arousals monitoring	Polysomnographic	92.80	High
Mousavi et al. [26]	Sleep stage detection	Single-channel EEG signals	93.55	High
Our proposed approach	Sleep stage recognition	EEG and ECG	93.76	Low
	Arousals detection	EEG and ECG	93.23	Low