Wearable Gait Monitoring With ML-Based Fall Risk Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for a cost-effective, non-obtrusive, and scalable technology for real-time gait monitoring that can predict and prevent falls by combining biomechanical and biometric data, provide real-time feedback, and overcome communication barriers for older adults, while being socially acceptable and convenient.
Innovation Solution
A method and system using wearable sensors and machine learning algorithms to monitor gait and posture, identifying fall risks through a smartphone or smartwatch, providing personalized feedback and emergency alerts, and adjusting to individual user behavior and preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex wearable systems with multiple sensors (IMUs, pressure, bend, height sensors) are used for gait monitoring, then measurement precision and reliability of fall detection are improved, but device complexity and obtrusiveness increase
Solution Approach 1:
The patent extracts and removes unnecessary complex sensor components from the wearable system. Instead of using multiple sensors (IMUs, pressure, bend, height sensors), the invention uses a simplified approach with fewer sensors, eliminating redundant elements while maintaining essential gait monitoring functionality for fall detection.
Solution Approach 2:
The patent makes the mobile device (smartphone or smartwatch) perform multiple functions: it serves as both the data collection platform and the processing unit for gait analysis. The mobile device's existing sensors are utilized for collecting gait data, and its processor analyzes the data to detect falls, eliminating the need for separate dedicated processing hardware.
2Reliability
If more sensors and advanced IoT technology are added for reliable gait analysis, then fall detection reliability is improved, but ease of operation and user comfort deteriorate
Solution Approach 1:
The patent utilizes the mobile device that users already possess and carry daily. This device serves multiple purposes including communication, entertainment, and now gait monitoring and fall detection. By leveraging the existing mobile device's sensors and processing capabilities, the system achieves reliable fall detection without adding separate wearable components that would reduce user comfort.
3Measurement precision
If traditional sensor-based systems are used for gait monitoring, then measurement capability is improved, but ease of manufacture and cost-effectiveness worsen
Solution Approach 1:
The patent leverages the mobile device's existing sensors (accelerometer, gyroscope, etc.) that are already integrated into smartphones and smartwatches. These sensors are mass-produced and inexpensive, eliminating the need for costly specialized sensor assemblies. The same sensors that users already have for other purposes are repurposed for gait analysis, significantly reducing manufacturing costs while maintaining measurement precision.
4Reliability
If real-time feedback and coaching systems are implemented, then fall prevention effectiveness is improved, but device complexity and power consumption increase
Solution Approach 1:
The patent utilizes the mobile device's existing processing unit and communication capabilities to provide real-time feedback and coaching. The processor analyzes gait data and generates feedback messages, while the device's display or notification system delivers the feedback to the user. This approach leverages the mobile device's already-present resources, avoiding the need for additional power-consuming feedback hardware.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A method and system for dynamic, non-obtrusive monitoring of locomotion of a person (50) using wearable sensor units (3) comprising motion sensors (6) arranged to generate a sensor signal (20), and a wearable communication unit (4) configured to process the sensor signal (20) using a signal processing unit (5) to extract and transmit biometric data (21) to a mobile device (1) that can analyze it using a machine learning-based risk prediction algorithm (40) to identify patterns (22) related to falling and thereby identify a fall event (23A) or calculate risk of falling (23B) of the person (50) .