Movement Analysis With Iterative Walking Boundary Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current devices inaccurately measure walking periods due to false readings from non-walking movements, particularly when worn on the wrist, affecting the validity of clinical assessments and treatment decisions.
Innovation Solution
A method and apparatus using an iterative optimization process to determine accurate start and stop times of walking or running periods by analyzing acceleration data from a motion sensor, employing autocorrelation functions and Hamming windows to refine epoch boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current movement measurement devices are used to detect walking periods, then the device can identify movement episodes, but false readings occur from non-walking movements such as wrist chopping activities
Solution Approach 1:
The system dynamically adjusts the walking detection criteria by continuously analyzing multiple parameters including acceleration magnitude, frequency content, and temporal patterns. The detection thresholds and parameters are adapted in real-time based on the user's movement characteristics, allowing the system to distinguish between walking and non-walking activities more accurately.
Solution Approach 2:
The patent changes multiple parameters simultaneously to improve detection accuracy: acceleration threshold levels, epoch duration, frequency analysis windows, and decision algorithms. By adjusting these parameters based on observed movement patterns, the system can differentiate between genuine walking episodes and false positive movements.
2Measurement precision
If current devices assume any vigorous movement constitutes the start of walking, then detection sensitivity is high, but the duration of walking is over-estimated
Solution Approach 1:
The system performs preliminary analysis of movement patterns before confirming the start of a walking episode. It pre-processes the acceleration data to identify potential walking onset points, then validates these candidates against multiple criteria including sustained acceleration patterns, frequency characteristics, and temporal consistency before finalizing the walking start time.
Solution Approach 2:
The system uses feedback from continuous analysis of acceleration patterns to refine walking episode detection. By monitoring the characteristics of detected movements and comparing them against expected walking patterns, the system can adjust its detection criteria and correct timing errors in real-time, reducing over-estimation of walking duration.
3Reliability
If activity monitoring is used for medical applications and clinical trials, then treatment decisions can be informed, but inaccurate activity data can nullify clinical assessments
Solution Approach 1:
The system is designed to serve multiple functions: it can detect various types of physical activities (walking, running, cycling), analyze different movement parameters, and adapt to different user populations. This multi-functionality allows the same device to be used across diverse clinical trials and medical applications, ensuring consistent and reliable data collection for treatment decision-making.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A movement analysis method and apparatus are disclosed. The apparatus and method receive acceleration data from an accelerometer (or measurements from other motion sensors) that is carried by a user. The acceleration data is divided into epochs and a determination is made as to whether the user is walking or running within each epoch. Consecutive epochs in which the user is found to be walking may be concatenated to determine periods of time during which the user is walking or running. An initial estimate of the time that the user starts to walk or run and stops walking or running are determined for each period of walking. An iterative optimisation process is then performed to determine accurate start and stop times for the period of walking or running.