Evaluating the feasibility of Fall detection using Single-channel EEG

Human Falls are a significant cause of injury and mortality worldwide, especially among older adults. Falls are the primary cause of mortality and fatal, non-fatal injuries in all age groups, according to the World Health Organization (WHO). Falls are the major cause of hospital admissions which impose substantial financial burdens on individuals, healthcare system, and society as a whole. True fall detection is a challenge in public health care. An automatic & accurate fall monitoring system is a must for fall detection & early assistance to reduce fall after-effects and has been a hot topic among researchers for the last two decades. Different techniques like vision, wearable, ambient, and muti-model are used for fall detection but the wearable technique is more suitable due to its cost-effectiveness & no area restriction on the subject. Most wearable techniques use accelerometer and gyroscope sensors whereas few research also going on muscular & cortical bioelectrical activity for fall detection and Brain-computer interface. This research presents the analysis of single-channel EEG signals for various non-fall and fall activities. This research aims to evaluate the feasibility of fall detection using morphological, statistical, and spectral analysis of EEG signals during non-fall and fall activities. The analysis shows significant variations in EEG signals for various non-fall and fall activities. The technique of single-channel EEG signal can be successfully used for discriminating fall events from non-fall events.