Related Experiment Video
Updated: Oct 6, 2025
08:05
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
10.9K
Human Fall Detection Using Passive Infrared Sensors with Low Resolution: A Systematic Review
Grégory Ben-Sadoun1,2, Emeline Michel3,4, Cédric Annweiler1,5,6,7
1Department of Geriatric Medicine and Memory Clinic, Research Center on Autonomy and Longevity, University Hospital of Angers, Angers, France.
Clinical Interventions in Aging
|January 20, 2022
Summary
Low-resolution passive infrared sensors show high effectiveness for human fall detection in older adults. Further real-world testing is recommended for these Information and Communication Technologies (ICTs) systems.
Area of Science:
- Gerontology
- Biomedical Engineering
- Computer Science
Background:
- Information and Communication Technologies (ICTs) offer solutions for human fall detection in older adults, balancing effectiveness and ethics.
- Low-resolution passive infrared (PIR) sensors present a potential technology for this application.
Purpose of the Study:
- To systematically review the metrological qualities of low-resolution (≤16x16 pixels) PIR sensors for fall detection.
- To assess the effectiveness of PIR sensor systems in identifying falls in older adults.
Main Methods:
- Systematic literature review of studies published until November 2020, sourced from PubMed, ScienceDirect, SpringerLink, IEEE Xplore, and MDPI.
- Analysis of 15 studies focusing exclusively on PIR sensors with a maximum resolution of 16x16 pixels for fall identification.
- Evaluation of study heterogeneity in experimental procedures and detection methods.
Main Results:
- Most studies (13/15) reported high performance metrics (accuracy, precision, sensitivity, or specificity) exceeding 85-90% for fall detection.
- Systems utilizing multiple sensors and advanced detection methods (e.g., 3D CNN, LSTM with CNN) achieved performance above 90%.
- Studies were predominantly conducted under controlled, empty-room conditions.
Conclusions:
- Low-resolution PIR sensor systems demonstrate promising effectiveness for human fall detection.
- Heterogeneity in study designs and testing environments limits definitive conclusions.
- Future research should prioritize evaluating these systems in real-life scenarios to ascertain their practical utility.
