Use of Hearing Aids Embedded with Inertial Sensors and Artificial Intelligence to Identify Patients at Risk for...

Area of Science:

• Gerontology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Fall risk assessment is crucial for preventing injuries in older adults.
• Traditional methods rely on trained observers and standardized tests.
• Emerging technologies like AI-powered hearing aids offer potential for objective fall risk evaluation.

Purpose of the Study:

• To compare fall risk scores derived from AI-powered hearing aids (IMU-HAs) with those obtained by trained clinicians.
• To evaluate the accuracy of IMU-HAs in categorizing fall risk using the STEADI test battery.

Main Methods:

• A prospective, double-blinded, observational study was conducted.
• 250 participants (aged 55-100) at risk for falls were enrolled.
• Fall risk was assessed using the STEADI tests (4-Stage Balance, TUG, 30-Second Chair Stand) via IMU-HAs and compared to clinician scores.

Main Results:

• Excellent interrater reliability was observed among clinicians.
• IMU-HAs demonstrated no significant difference from clinicians for 4-Stage Balance and TUG tests.
• IMU-HAs had a 12% failure rate in TUG trials and a significant difference in the 30-Second Chair Stand test, altering classification in 21% of participants.

Conclusions:

• AI-powered hearing aids largely align with clinician fall risk assessments.
• Discrepancies in specific tests highlight the need for algorithmic refinement.
• Further development is required to ensure accurate fall risk categorization by IMU-HAs.