Related Experiment Video
Updated: Jan 9, 2026
Computerized Dynamic Posturography for Postural Control Assessment in Patients with Intermittent Claudication
Published on: December 11, 2013
AI-Assisted Dynamic Postural Control Screening to Improve Functional Mobility in Older Adult Populations:
Kai-Chih Lin1, Rong-Jong Wai1, Hung-Yu Chang Chien2
1Department of Electronic and Computer Engineering, National Taiwan University of Science and Technology, No. 43, Sec. 4, Keelung Rd., Da'an Dist., Taipei, 106335, Taiwan, 886 935636806, 886 227376367.
Background:
Falls are a major cause of disability among older adults, and early identification of functional decline is essential for prevention. Artificial intelligence (AI) systems may enhance mobility screening by providing objective, real-time feedback.
Objective:
This study aimed to evaluate whether AI-assisted dynamic postural control screening combined with adaptive training improves functional mobility outcomes in older adult populations.
Methods:
A quasi-experimental study was conducted with 2005 older adults recruited from community centers and health care institutions in Keelung, Taiwan. Participants were assigned to either an experimental group (n=1451), which underwent AI-assisted screening with adaptive exercise prescriptions, or a control group (n=554), which completed follow-ups through regular physical assessments with standard care without AI-tailored training. The AI system integrated skeletal tracking with the Short Physical Performance Battery to assess balance, gait speed (4-m walk), and sit-to-stand performance. Independent-samples 2-tailed t tests and repeated-measures ANOVA were applied, and effect sizes (Cohen d and η²) with 95% CIs were reported.
Results:
The experimental group demonstrated significantly greater improvements compared with the control group in Short Physical Performance Battery scores (Δ=0.8 vs 0.3; t2003=3.41; P=.001; Cohen d=0.45, 95% CI 0.18-0.72), gait speed (Δ=15 cm/s vs 5 cm/s; t2003=4.85; P<.001; Cohen d=0.62, 95% CI 0.35-0.88), and sit-to-stand time (Δ=-1.4 s vs -0.6 s; t2003=3.12; P=.002; Cohen d=0.39, 95% CI 0.12-0.65). Here "Δ" refers to the change score, calculated as post-intervention minus baseline (ie, the amount of improvement during the study period). Participation rate was strongly associated with outcomes, with 1-way ANOVA showing significant group differences (F2,1448=8.74-12.21; P<.001; η²=0.07-0.10).
Conclusions:
AI-assisted dynamic postural control screening combined with adaptive training substantially improved functional performance in mobility, balance, and gait among older adults. While fall incidence was not directly measured, these functional gains may have implications for fall risk reduction. Future longitudinal studies with extended follow-up (12-24 mo) and prospective fall incidence tracking across diverse populations are required to validate whether these improvements translate into actual reductions in fall risk.
