Wearable Fall Detection Using Audio and IMU Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional personal emergency response systems (PERS) face challenges in accurately detecting falls and estimating user pain levels, often resulting in false positives and insufficient assessment of fall-related pain, particularly in older populations with cognitive impairments, and require more efficient methods for monitoring physical activity and wear detection.
Innovation Solution
A multimodal approach incorporating auditory, accelerometric, and heart-rate-based detection using audio sensors and inertial measurement units (IMUs) to enhance fall detection accuracy, estimate fall intensity, and assess pain levels through post-fall sound signals and heart rate variability, enabling more precise and user-aware urgent care.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional accelerometric detection is used for fall detection, then the system can detect fall events, but false positives occur and pain level assessment is insufficient
Solution Approach 1:
The patent combines multiple detection modalities (accelerometric data from IMUs, audio data from microphones, and ECG data from electrodes) into a unified fall detection and pain assessment system. This multi-sensor fusion approach resolves the contradiction by using accelerometric data for fall event detection while simultaneously using audio and physiological data for accurate pain level assessment, eliminating the insufficiency of single-modal detection.
Solution Approach 2:
The patent introduces audio signals as an intermediary to bridge fall detection and pain assessment. Post-fall audio recordings capture verbal expressions of pain, which are processed to determine pain levels. This intermediary approach allows the system to accurately assess pain without relying solely on accelerometric data, resolving the measurement precision issue while maintaining reliable fall detection.
2Quantity of substance
If PERS monitors are worn continuously for physical activity measurement, then activity data can be collected, but it is difficult to differentiate between no activity while wearing versus not wearing the monitor
Solution Approach 1:
The patent uses audio detection as an intermediary to determine whether the PERS monitor is being worn. The system attempts to detect characteristic audio patterns (such as ambient sounds or user interactions) that indicate the presence of the device on the user's body. This resolves the contradiction by providing a separate verification mechanism that distinguishes between non-wear and wear-with-no-activity scenarios.
Solution Approach 2:
The PERS monitor is designed with multi-functionality, serving both as a fall detection device and a wear detection device. The same audio sensors and processing capabilities used for fall verification are also employed to detect whether the device is being worn, eliminating the need for separate detection mechanisms and enabling accurate differentiation between wear states.
3Measurement precision
If ECG monitors are attached to the user's chest for heart rate measurement, then accurate heart rate data can be obtained, but it is inconvenient for long-term monitoring
Solution Approach 1:
The patent replaces the traditional chest-attached ECG monitor with a disposable or removable adhesive patch containing electrodes. This patch can be easily applied and removed by the user, providing accurate ECG data for heart rate measurement without the inconvenience of permanent chest attachment. The simplified application process maintains measurement precision while dramatically improving ease of operation for long-term monitoring.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method are provided for performing physical activity monitoring and fall detection of a subject using a wearable sensor including at least one audio sensor and at least one of an accelerometer, a gyroscope and a magnetometer. The method includes monitoring activities of the subject using the at least one of the accelerometer, the gyroscope and the magnetometer, and identifying a characteristic motion pattern based on the monitored activities; detecting an apparent fall experienced by the subject based on the identified at least one characteristic motion pattern; monitoring sounds provided by the at least one audio sensor following the detected apparent fall; determining whether the detected apparent fall is an actual fall experienced by the subject based on the monitored sounds; and communicating an indication of the actual fall to a monitoring system, enabling commencement of responsive action.