Automated fall risk assessment of elderly using wearable devices

Area of Science:

• Gerontology
• Biomedical Engineering
• Rehabilitation Science

Background:

• Falls represent a significant financial burden on the healthcare system.
• Current fall risk assessments are often labor-intensive and prone to human error.

Purpose of the Study:

• To investigate the efficacy of wearable devices and algorithms for automated fall risk assessment.
• To compare automated test results with observational assessments by physiotherapists.

Main Methods:

• 13 participants completed standardized fall risk tests (6-Minutes Walk, Timed-Up-and-Go, 30-Second Sit-to-Stand, 4-Stage Balance).
• Wearable device algorithms and visual data analysis were compared against physiotherapist observations.
• A total of 226 tests were analyzed.

Main Results:

• High congruence (78.15%–96.55%) was observed between automated assessments and ground truth for all tests.
• Deviations were within one standard deviation of the ground truth.
• Fall risk, assessed via questionnaire, correlated with individual test results.

Conclusions:

• Automated fall risk assessment using wearable technology is a valid and resourceful alternative to observational methods.
• This approach minimizes human error in fall risk evaluation.
• Predicting overall fall risk requires a complex model incorporating multiple parameters beyond single test results.