A two-layer hierarchical framework for activity sequence recognition by wearable sensors

Recognition of Bathroom Activities in Older Adults Using Wearable Sensors: A Systematic Review and Recommendations

“…In total, seven distinct bathroom activities were investigated, which can be listed according to the number of times they were included: toileting [6,9,24,27], brushing teeth [7,25,26], washing hands [6,7], combing hair [7], dressing [24] and using a bathroom sink [9]. Therefore, this review study shows that most frequently studied bathroom activities are those characterised by simple movements, like toileting, or those characterised by periodic movements, like brushing teeth.…”

“…In this systematic review, the focus was on wearable sensors, attached to different body positions. Sensors were attached to the participant's hands [25,26], to the wrist [6,9], to the waist [6,9], to the chest [7], to the lower back [9] and to the thigh [9]. One study integrated the sensors in a toothbrush, as it was only focused on different types of brushing teeth patterns [27].…”

“…One prominent element is the decreased ability of performing Activities of Daily Living (ADL), which affects older adults' ability to live independently at home. In the past years, a variety of devices have been used and researched to overcome limitations related to ageing and to help older adults stay independently at home, including wearable sensors, cameras, audio recorders and Internet of Things devices [6][7][8][9][10]. These devices can be used for recognising specific situations and activities in the daily life of older adults.…”

“…However, to the best of the authors' knowledge, few studies have been conducted on bathroom activity monitoring for older adults, especially using wearable sensors [6,7,9,[24][25][26][27]. Therefore, the objective of this study is to review such studies and examine the actual studied bathroom activities, used wearable sensors, different applied methodologies and the activity recognition techniques.…”

“…When activities have similar movements like washing hands, or washing food, environmental or physiological sensors can be used for detecting (previously conducted) related activities. In five of the seven included studies, the experimental procedures were not described in detail [6,9,[25][26][27]. Therefore, it is hard to evaluate their experimental protocol and impossible for others to repeat their experiments, making it hard to generalise results.…”