Foot-worn Fall Prediction Using Force and IMU Data Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for predicting falls using sensors on a subject's foot lack effectiveness in accurately identifying imminent falls and providing timely alerts, as they do not adequately integrate multiple data sources and machine learning algorithms to analyze complex gait and movement patterns.
Innovation Solution
A method and system that utilize a combination of force sensors and inertial measurement units (IMUs) worn on the foot, integrated with a machine learning model that includes deep learning neural networks and random forest algorithms, to analyze force and IMU data, along with supplemental data, to generate fall prediction data and provide alerts based on risk scores and contributors to falls.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple data sources and machine learning algorithms are integrated to analyze complex gait and movement patterns, then fall prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The system segments the fall prediction functionality into distinct modular components: force sensors for vertical force measurement, IMUs for motion tracking, and separate machine learning models for different analysis tasks. This modular architecture improves prediction accuracy through comprehensive data collection while managing complexity through organized, independent modules that can be developed and maintained separately.
Solution Approach 2:
The system merges multiple data sources (force sensors and IMUs) and multiple machine learning algorithms into an integrated fall prediction system. The force data and IMU data are combined and processed together through coordinated machine learning models, achieving improved fall prediction accuracy by leveraging the complementary strengths of different sensing modalities and analytical approaches.
2Reliability
If multiple sensors and algorithms are used to provide timely alerts, then fall detection reliability is improved, but processing time increases
Solution Approach 1:
The machine learning models are trained in advance on extensive gait and movement data to pre-learn patterns associated with fall risk. During operation, the system leverages this pre-trained knowledge to rapidly analyze incoming sensor data and generate alerts, achieving high reliability through comprehensive analysis while minimizing processing time by avoiding the need for complex real-time computations.
Data Source
AI summary
A method for predicting a fall includes: obtaining a series of force data using a plurality of force sensors worn on a subject's foot over a time window; obtaining a series of inertial measurement unit (IMU) data from at least a first IMU worn on the subject's foot over the time window; inputting input data into a machine learning model, where the input data includes at least the series of force data and the series of IMU data, and where the machine learning model is trained to employ a set of parameters to generate fall prediction data using the input data; and outputting an indication of the fall prediction data.