Gait Assessment Using Depth Cameras and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The subjective nature of the Unified Parkinson's Disease Rating Scale's Postural Instability and Gait Disturbance test makes it difficult for medical professionals to objectively assess gait in patients with Parkinson's disease, leading to poor correlation and consistency between assessments.
Innovation Solution
A system and method using machine learning models, inertial sensors, and data from various sources to objectively assess gait by collecting and processing data on tremors, movement, muscular rigidity, and other gait-related features, with feedback loops to improve model accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the Unified Parkinson's Disease Rating Scale's Postural Instability and Gait Disturbance test is used, then gait assessment can be performed, but the assessment becomes subjective and lacks consistency between different medical professionals
Solution Approach 1:
The patent replaces the manual mechanical assessment process with an automated computer vision system. Depth cameras capture gait data, and machine learning algorithms automatically analyze movement patterns, substituting the subjective mechanical evaluation with an objective digital system that eliminates inter-profiler variability.
Solution Approach 2:
The system creates a digital copy of the patient's gait through depth camera imaging and machine learning modeling. This digital representation captures movement characteristics without physical contact, allowing consistent reproduction of the assessment across different users and time points without the variability inherent in manual evaluation.
2Reliability
If manual gait assessment is performed by medical professionals, then diagnostic information can be obtained, but the process is time-consuming and lacks objectivity
Solution Approach 1:
The system enables self-service assessment where patients can undergo gait evaluation without requiring medical professionals present. The automated machine learning system performs the diagnostic assessment independently, providing objective results that can be obtained in the patient's own environment without time-consuming clinical visits.
Solution Approach 2:
The depth camera system enables continuous or repeated gait assessments without interruption. Unlike manual assessment requiring scheduled appointments, the automated system can operate continuously, providing ongoing monitoring and consistent diagnostic information over time without the time losses associated with repeated clinical visits.
3Measurement precision
If objective gait assessment tools are developed, then measurement consistency can be improved, but the device complexity and data processing requirements increase
Solution Approach 1:
The machine learning model is designed to handle multiple gait parameters and disease stages through a single unified system. The same depth camera and algorithm framework can assess various gait characteristics (stride length, speed, symmetry) and detect different Parkinson's disease severities, eliminating the need for multiple specialized devices and reducing overall system complexity.
Solution Approach 2:
The system dynamically adjusts measurement parameters based on the patient's gait characteristics and disease severity. The machine learning algorithm adapts its analysis based on the data collected, automatically optimizing measurement parameters for each individual case without requiring manual configuration or complex pre-programmed settings, thereby reducing system complexity while maintaining high measurement precision.
Data Source
AI summary
The exemplary embodiments disclose a system and method, a computer program product, and a computer system for assessing a user's gate. The exemplary embodiments may include collecting data corresponding to a walking user and assessing a gait of the user based on applying one or more models to the data.