Background: Social isolation during COVID-19 may negatively impact older adults’ wellbeing. To assess its impact, we measured changes in physical activity and sleep among community-dwelling older adults, from pre-to post-pandemic declaration. Method: Physical activity and sleep in older adults (n = 10, age = 77.3 ± 1.9 years, female = 40%) were remotely assessed within 3-month pre-to 6-month post-pandemic declaration using a pendant-wearable system. Depression was assessed pre-and post-pandemic declaration using the Center for Epidemiologic Studies Depression scale and was compared with 48 h continuous physical activity monitoring data before and during pandemic. Results: Compared to pre-pandemic, post-pandemic time spent in standing declined by 32.7% (Cohen’s d = 0.78, p < 0.01), walking by 52.2% (d = 1.1, p < 0.01), step-counts by 55.1% (d = 1.0, p = 0.016), and postural transitions by 44.6% (d = 0.82, p = 0.017) with increase in sitting duration by 20.5% (d = 0.5, p = 0.049). Depression symptoms increased by 150% (d = 0.8, p = 0.046). Interestingly, increase in depression was significantly correlated with unbroken-prolong sitting bout (ρ = 0.677, p = 0.032), cadence (ρ = −0.70, p = 0.024), and sleep duration (ρ = −0.72, p = 0.019). Conclusion: This is one of the early longitudinal studies highlighting adverse effect of the pandemic on objectively assessed physical activity and sleep in older adults. Our observations showed need for timely intervention to mitigate hard to reverse consequences of decreased physical activity such as depression.

Remote monitoring of physical frailty is important to personalize care for slowing down the frailty process and/or for the healthy recovery of older adults following acute or chronic stressors. Taking the Fried frailty criteria as a reference to determine physical frailty and frailty phenotypes (slowness, weakness, exhaustion, inactivity), this study aimed to explore the benefit of machine learning to determine the least number of digital biomarkers of physical frailty measurable from a pendant sensor during activities of daily living. Two hundred and fifty-nine older adults were classified into robust or pre-frail/frail groups based on the physical frailty assessments by the Fried frailty criteria. All participants wore a pendant sensor at the sternum level for 48 h. Of seventeen sensor-derived features extracted from a pendant sensor, fourteen significant features were used for machine learning based on logistic regression modeling and a recursive feature elimination technique incorporating bootstrapping. The combination of percentage time standing, percentage time walking, walking cadence, and longest walking bout were identified as optimal digital biomarkers of physical frailty and frailty phenotypes. These findings suggest that a combination of sensor-measured exhaustion, inactivity, and speed have potential to screen and monitor people for physical frailty and frailty phenotypes.

Improved life expectancy is increasing the number of older adults who suffer from motor-cognitive decline. Unfortunately, conventional balance exercise programs are not tailored to patients with cognitive impairments, and exercise adherence is often poor due to unsupervised settings. This study describes the acceptability and feasibility of a sensor-based in-home interactive exercise system, called tele-Exergame, used by older adults with mild cognitive impairment (MCI) or dementia. Our tele-Exergame is specifically designed to improve balance and cognition during distractive conditioning while a telemedicine interface remotely supervises the exercise, and its exercises are gamified balance tasks with explicit augmented visual feedback. Fourteen adults with MCI or dementia (Age = 68.1 ± 5.4 years, 12 females) participated and completed exergame twice weekly for six weeks at their homes. Before and after 6 weeks, participants’ acceptance was assessed by Technology Acceptance Model (TAM) questionnaire, and participants’ cognition and anxiety level were evaluated by the Montreal Cognitive Assessment (MoCA) and Beck Anxiety Inventory (BAI), respectively. Results support acceptability, perceived benefits, and positive attitudes toward the use of the system. The findings of this study support the feasibility, acceptability, and potential benefit of tele-Exergame to preserve cognitive function among older adults with MCI and dementia.

Since conventional screening tools for assessing frailty phenotypes are resource intensive and unsuitable for routine application, efforts are underway to simplify and shorten the frailty screening protocol by using sensor-based technologies. This study explores whether machine learning combined with frailty modeling could determine the least sensor-derived features required to identify physical frailty and three key frailty phenotypes (slowness, weakness, and exhaustion). Older participants (n = 102, age = 76.54 ± 7.72 years) were fitted with five wearable sensors and completed a five times sit-to-stand test. Seventeen sensor-derived features were extracted and used for optimal feature selection based on a machine learning technique combined with frailty modeling. Mean of hip angular velocity range (indicator of slowness), mean of vertical power range (indicator of weakness), and coefficient of variation of vertical power range (indicator of exhaustion) were selected as the optimal features. A frailty model with the three optimal features had an area under the curve of 85.20%, a sensitivity of 82.70%, and a specificity of 71.09%. This study suggests that the three sensor-derived features could be used as digital biomarkers of physical frailty and phenotypes of slowness, weakness, and exhaustion. Our findings could facilitate future design of low-cost sensor-based technologies for remote physical frailty assessments via telemedicine.

Aggression in children is highly prevalent and can have devastating consequences, yet there is currently no objective method to track its frequency in daily life. This study aims to investigate the use of wearable-sensor-derived physical activity data and machine learning to objectively identify physical-aggressive incidents in children. Participants (n = 39) aged 7 to 16 years, with and without ADHD, wore a waist-worn activity monitor (ActiGraph, GT3X+) for up to one week, three times over 12 months, while demographic, anthropometric, and clinical data were collected. Machine learning techniques, specifically random forest, were used to analyze patterns that identify physical-aggressive incident with 1-min time resolution. A total of 119 aggression episodes, lasting 7.3 ± 13.1 min for a total of 872 1-min epochs including 132 physical aggression epochs, were collected. The model achieved high precision (80.2%), accuracy (82.0%), recall (85.0%), F1 score (82.4%), and area under the curve (89.3%) to distinguish physical aggression epochs. The sensor-derived feature of vector magnitude (faster triaxial acceleration) was the second contributing feature in the model, and significantly distinguished aggression and non-aggression epochs. If validated in larger samples, this model could provide a practical and efficient solution for remotely detecting and managing aggressive incidents in children.

Background: A critical factor in healing diabetic foot ulcers is patient adherence to offloading devices. We tested a smart offloading boot (SmartBoot) combined with a smartwatch app and cloud dashboard to remotely monitor patient adherence and activity. In addition, the impact of SmartBoot on balance, gait, and user experience was investigated. Methods: Fourteen volunteers (31.6±8.7 years; 64% female) performed natural activities (eg, sitting, standing, walking) with and without the SmartBoot for approximately 30 minutes. All participants completed balance tests, 10-meter walking tests at slow, normal, and fast pace while wearing the SmartBoot, and a user experience questionnaire. The accuracy of real-time adherence reporting was assessed by comparing the SmartBoot and staff observation. Center of mass (COM) sway and step counts were measured using a validated wearable system. Results: Average sensitivity, specificity, and accuracy for adherence and non-adherence were 90.6%, 88.0%, and 89.3%, respectively. The COM sway area was significantly smaller with the SmartBoot than without the SmartBoot regardless of test condition. Step count error was 4.4% for slow waking, 36.2% for normal walking, 16.0% for fast walking. Most participants agreed that the SmartBoot is easy to use, relatively comfortable, nonintrusive, and innovative. Conclusions: To our knowledge, this is the first smart offloading system that enables remote patient monitoring and real-time adherence and activity reporting. The SmartBoot enhanced balance performance, likely due to somatosensory feedback. Questionnaire results highlight SmartBoot’s technical and clinical potential. Future studies warrant clinical validation of real-time non-adherence alerting to improve wound healing outcomes in people with diabetic foot ulcers.