Towards safer mobility: Developing and evaluating a fall detection system for a smart walker

Cavaglià, Maria Sole; Sierra M, Sergio D.; Palmerini, Luca; Orlandi, Silvia; Múnera, Marcela; Cifuentes, Carlos A.

Abstract

Fall detection and prevention is a key issue for healthcare in older adults since it prevents the development of multiple cognitive and physical disorders. This study aims to evaluate multiple falls and near-fall detection algorithms to be integrated into a smart walker, which usually comes with an increased risk for falls in case of improper use. A six-axis IMU worn at the user's waist extracts trunk inclination, angular velocity and acceleration. The study employs various fall detection algorithms, such as Kangas and Vertical Velocity, tailored for fall detection, Triangular Feature, Vertical Angle and a Modified Vertical Velocity for pre-fall detection. The experimental protocol involved various Activities of Daily Living (ADLs) and simulated falls, emphasizing participant safety and data usability. The results from this study provide insights into the effectiveness and reliability of the integrated fall detection system in scenarios involving a smart walker.

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2024 10th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob)
• Start Date: Sep 1, 2024
• End Date: Sep 4, 2024
• Acceptance Date: Apr 30, 2024
• Online Publication Date: Oct 23, 2024
• Publication Date: Oct 23, 2024
• Deposit Date: Jul 2, 2024
• Publicly Available Date: Oct 23, 2024
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Pages: 1623-1628
• Series ISSN: 2155-1782
• Book Title: 2024 10th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob)
• DOI: https://doi.org/10.1109/BioRob60516.2024.10719869
• Public URL: https://uwe-repository.worktribe.com/output/12040497