Fall Detector Using Physiological Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fall detection systems face challenges in accurately differentiating between fall and non-fall events, especially when sensors are located on areas like the wrist, leading to increased false alarms and reduced adherence due to overlapping parameter values for motion measurements.
Innovation Solution
The integration of physiological measurements such as heart rate and skin conductance data, which respond uniquely to falls, is used in conjunction with traditional motion data to enhance the distinguishing power of the fall detection system, employing machine learning classifiers to optimize feature extraction and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion sensors (accelerometers, gyroscopes, magnetometers) are used for fall detection, then the system can detect motion parameters, but the accuracy decreases when sensors are located on areas like the wrist due to overlapping parameter values between fall and non-fall events
Solution Approach 1:
The patent adds physiological measurement dimensions (heart rate, skin conductance) to the existing motion measurement dimensions. This multi-dimensional approach allows the system to differentiate between fall and non-fall events by analyzing patterns across multiple parameter spaces simultaneously, resolving the overlap problem that occurs when using motion sensors alone on peripheral body locations.
2Measurement precision
If physiological sensors (heart rate, skin conductance) are added to enhance fall detection accuracy, then the distinguishing power between fall and non-fall events improves, but the device complexity and number of sensors increase
Solution Approach 1:
The patent merges motion sensing capabilities with physiological sensing capabilities into a unified fall detection system. By combining data from accelerometers, gyroscopes, magnetometers, heart rate sensors, and skin conductance sensors, the system creates a comprehensive measurement model that leverages the complementary strengths of each sensor type to achieve high accuracy without requiring any single sensor to be perfectly positioned.
Solution Approach 2:
The fall detection system is designed to operate effectively regardless of sensor location on the body. The multi-sensor approach allows the system to function as a universal fall detector that can be worn on the wrist, torso, or other locations, adapting to different user preferences and comfort levels while maintaining detection accuracy through its ability to analyze multiple parameter dimensions simultaneously.
3Measurement precision
If machine learning classifiers are used to process multiple sensor inputs, then the accuracy of fall detection improves, but the computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary processing of sensor data by extracting relevant features from raw accelerometer, gyroscope, magnetometer, heart rate, and skin conductance signals before feeding them to the machine learning classifier. This preprocessing step reduces the dimensionality and complexity of the input data, enabling more efficient classification while maintaining high accuracy in distinguishing fall from non-fall events.
Data Source
AI summary
A method for detecting a fall by a user wearing a fall detector, including: detecting a trigger event identifying the time location of a possible fall event in user data; extracting motion features from motion data and physiological features from physiological data from within a time window around the identified time location; and determining whether the detected trigger event is a fall by the user by inputting the at least one of the motion features and at least one of the physiological features into a classifier.