SWORD - South West Open Research Deposit - International Symposium of Adapted Physical Activity and International Symposium on Physical Activity and Visual Impairment and Deafblindness: AI-Driven Adapted Physical Activity to Promote Health and Mobility for Frail Older Adults
 

Start Date

19-6-2025 10:30 AM

End Date

19-6-2025 12:00 PM

Abstract

Introduction Physical activity (PA) programs have been widely implemented to promote health and maintain functional mobility in older adults. However, existing PA programs are mainly single component (e.g., resistance training, walking) and do not consider the diverse functional capacities of older adults. This study aimed to develop and evaluate an artificial intelligence (AI)-assisted PA program for enhancing mobility in frail older adults.

Methodology We developed a computer algorithm that can generate PA prescription based on the functional status of older adults and a multi-component PA program, Mobility-Fit (Yang et al. 2024). 146 older adults from 20 care facilities were recruited and randomly assigned into the intervention group (practicing the adapted PA for 12 weeks) or a control group that maintained usual care. Participants underwent assessments of strength, mobility, cognition, PA level, and falls efficacy. Baseline measurements were incorporated into the computer algorithm to generate customized recommendations to guide the PA delivery. Repeated-Measures Linear Models were used to examine changes within and between groups.

Results 112 participants completed the study (age: 84.4±8.0). Among the intervention group, handgrip strength increased by 15% (p=0.036), knee extension strength increased by 16% (p=0.012), reaction time decreased by 7% (p=0.049), and five-time sit-to-stand duration decreased by 12% (p=0.035). Significant interactions were found between time and group in these variables (p<0.05). The adherence to PA practice was 75.4%.

Conclusions Our AI-driven adapted PA program effectively improved strength and mobility, which holds the potential to enhance efficiency of healthcare practitioners by providing customized PA for frail older adults.

Share

COinS
 
Jun 19th, 10:30 AM Jun 19th, 12:00 PM

AI-Driven Adapted Physical Activity to Promote Health and Mobility for Frail Older Adults

Introduction Physical activity (PA) programs have been widely implemented to promote health and maintain functional mobility in older adults. However, existing PA programs are mainly single component (e.g., resistance training, walking) and do not consider the diverse functional capacities of older adults. This study aimed to develop and evaluate an artificial intelligence (AI)-assisted PA program for enhancing mobility in frail older adults.

Methodology We developed a computer algorithm that can generate PA prescription based on the functional status of older adults and a multi-component PA program, Mobility-Fit (Yang et al. 2024). 146 older adults from 20 care facilities were recruited and randomly assigned into the intervention group (practicing the adapted PA for 12 weeks) or a control group that maintained usual care. Participants underwent assessments of strength, mobility, cognition, PA level, and falls efficacy. Baseline measurements were incorporated into the computer algorithm to generate customized recommendations to guide the PA delivery. Repeated-Measures Linear Models were used to examine changes within and between groups.

Results 112 participants completed the study (age: 84.4±8.0). Among the intervention group, handgrip strength increased by 15% (p=0.036), knee extension strength increased by 16% (p=0.012), reaction time decreased by 7% (p=0.049), and five-time sit-to-stand duration decreased by 12% (p=0.035). Significant interactions were found between time and group in these variables (p<0.05). The adherence to PA practice was 75.4%.

Conclusions Our AI-driven adapted PA program effectively improved strength and mobility, which holds the potential to enhance efficiency of healthcare practitioners by providing customized PA for frail older adults.