Systematic Review of Single-Channel EEG-Based Drowsiness Detection Methods

Abstract

Drowsiness is characterized by reduced attentiveness, commonly experienced during the transition from wakefulness to sleepiness. It can decrease an individual’s alertness, thereby increasing the risk of accidents during activities such as driving, crane operation, working in mining areas, and industrial machinery operation. The detection of drowsiness plays an important role in preventing such accidents. Numerous methods exist for drowsiness detection, including Subjective, Vehicle, Behavioral, and Physiological approaches. Among these, Physiological methods, particularly those utilizing Electroencephalogram (EEG) data combined with artificial intelligence, have proven effective in detecting drowsiness. These methods excel in capturing physiological changes in the body during drowsiness and the potential for gathering information from the human brain during this state. EEG data-based Brain-Computer interface (BCI) systems have been popular for detecting drowsiness. Single-channel EEG signal analysis BCIs have been highly preferred for their ease and convenient usage in real-time applications. While some progress has been made in the single-channel EEG BCI, substantial progress is still needed. This paper provides a state-of-the-art analysis of recent developments in the single-channel EEG-based drowsiness detection methods. Ultimately, this review study explores potential avenues for the future development of single-channel EEG-based drowsiness detection.