A Night-Time Monitoring System (eNightLog) to Prevent Elderly Wandering in Hostels: A Three-Month Field Study.

Older people are increasingly dependent on others to support their daily activities due to geriatric symptoms such as dementia. Some of them stay in long-term care facilities. Elderly people with night wandering behaviour may lose their way, leading to a significant risk of injuries. The eNightLog system was developed to monitor the night-time bedside activities of older people in order to help them cope with this issue. It comprises a 3D time-of-flight near-infrared sensor and an ultra-wideband sensor for detecting human presence and to determine postures without a video camera. A threshold-based algorithm was developed to classify different activities, such as leaving the bed. The system is able to send alarm messages to caregivers if an elderly user performs undesirable activities. In this study, 17 sets of eNightLog systems were installed in an elderly hostel with 17 beds in 9 bedrooms. During the three-month field test, 26 older people with different periods of stay were included in the study. The accuracy, sensitivity and specificity of detecting non-assisted bed-leaving events was 99.8%, 100%, and 99.6%, respectively. There were only three false alarms out of 2762 bed-exiting events. Our results demonstrated that the eNightLog system is sufficiently accurate to be applied in the hostel environment. Machine learning with instance segmentation and online learning will enable the system to be used for widely different environments and people, with improvements to be made in future studies.

Swallow Detection with Acoustics and Accelerometric-Based Wearable Technology: A Scoping Review.

Swallowing disorders, especially dysphagia, might lead to malnutrition and dehydration and could potentially lead to fatal aspiration. Benchmark swallowing assessments, such as videofluoroscopy or endoscopy, are expensive and invasive. Wearable technologies using acoustics and accelerometric sensors could offer opportunities for accessible and home-based long-term assessment. Identifying valid swallow events is the first step before enabling the technology for clinical applications. The objective of this review is to summarize the evidence of using acoustics-based and accelerometric-based wearable technology for swallow detection, in addition to their configurations, modeling, and assessment protocols. Two authors independently searched electronic databases, including PubMed, Web of Science, and CINAHL. Eleven (n = 11) articles were eligible for review. In addition to swallowing events, non-swallowing events were also recognized by dry (saliva) swallowing, reading, yawning, etc., while some attempted to classify the types of swallowed foods. Only about half of the studies reported that the device attained an accuracy level of >90%, while a few studies reported poor performance with an accuracy of <60%. The reviewed articles were at high risk of bias because of the small sample size and imbalanced class size problem. There was high heterogeneity in assessment protocol that calls for standardization for swallowing, dry-swallowing and non-swallowing tasks. There is a need to improve the current wearable technology and the credibility of relevant research for accurate swallowing detection before translating into clinical screening for dysphagia and other swallowing disorders.

A Blanket Accommodative Sleep Posture Classification System Using an Infrared Depth Camera: A Deep Learning Approach with Synthetic Augmentation of Blanket Conditions.

Surveillance of sleeping posture is essential for bed-ridden patients or individuals at-risk of falling out of bed. Existing sleep posture monitoring and classification systems may not be able to accommodate the covering of a blanket, which represents a barrier to conducting pragmatic studies. The objective of this study was to develop an unobtrusive sleep posture classification that could accommodate the use of a blanket. The system uses an infrared depth camera for data acquisition and a convolutional neural network to classify sleeping postures. We recruited 66 participants (40 men and 26 women) to perform seven major sleeping postures (supine, prone (head left and right), log (left and right) and fetal (left and right)) under four blanket conditions (thick, medium, thin, and no blanket). Data augmentation was conducted by affine transformation and data fusion, generating additional blanket conditions with the original dataset. Coarse-grained (four-posture) and fine-grained (seven-posture) classifiers were trained using two fully connected network layers. For the coarse classification, the log and fetal postures were merged into a side-lying class and the prone class (head left and right) was pooled. The results show a drop of overall F1-score by 8.2% when switching to the fine-grained classifier. In addition, compared to no blanket, a thick blanket reduced the overall F1-scores by 3.5% and 8.9% for the coarse- and fine-grained classifiers, respectively; meanwhile, the lowest performance was seen in classifying the log (right) posture under a thick blanket, with an F1-score of 72.0%. In conclusion, we developed a system that can classify seven types of common sleeping postures under blankets and achieved an F1-score of 88.9%.

Night-Time Monitoring System (eNightLog) for Elderly Wandering Behavior.

Wandering is a common behavioral disorder in the community-dwelling elderly. More than two-thirds of caregivers believe that wandering would cause falls. While physical restraint is a common measure to address wandering, it could trigger challenging behavior in approximately 80% of the elderly with dementia. This study aims to develop a virtual restraint using a night monitoring system (eNightLog) to provide a safe environment for the elderly and mitigate the caregiver burden. The eNightLog system consisted of remote sensors, including a near infra-red 3D time-of-flight sensor and ultrawideband sensors. An alarm system was controlled by customized software and algorithm based on the respiration rate and body posture of the elderly. The performance of the eNightLog system was evaluated in both single and double bed settings by comparing to that of a pressure mat and an infrared fence system, under simulated bed-exiting scenarios. The accuracy and precision for the three systems were 99.0%, 98.8%, 85.9% and 99.2%, 97.8%, 78.6%, respectively. With higher accuracy, precision, and a lower false alarm rate, eNightLog demonstrated its potential as an alternative to physical restraint to remedy the workload of the caregivers and the psychological impact of the elderly.