Locogram Cycle Similarity Matrix for Gait Regularity and Symmetry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gait analysis methods using inertial sensors rely heavily on parameters that average gait characteristics over the entire exercise, obscuring progress and failing to provide analytical information about the origin of changes in gait, and are affected by inter-individual variability without a reference method to overcome this issue.
Innovation Solution
A visualization matrix, or 'locogram', is constructed to compare each walking or running cycle with others, using similarity coefficients to account for the shape of the signal, allowing for a visual representation that highlights irregularities and symmetry, independent of sensor type or anatomical location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If parameters are used to summarize gait characteristics, then the overall gait performance can be assessed, but the temporal progression and origin of gait changes are obscured
Solution Approach 1:
The gait signal is segmented into individual cycles, and each cycle is compared with every other cycle to generate a matrix of similarity coefficients. This segmentation preserves temporal information by maintaining the identity of each cycle while enabling comprehensive comparison across the entire gait sequence.
Solution Approach 2:
The patent transforms the one-dimensional time series gait signal into a two-dimensional matrix where rows and columns represent different cycles and cells represent similarity coefficients. This dimensional transformation allows simultaneous visualization of temporal progression and cycle-to-cycle variations.
2Productivity
If thresholds are defined to calculate parameters, then gait metrics can be computed, but errors related to threshold selection occur
Solution Approach 1:
Instead of using fixed thresholds to define gait parameters, the patent transforms the problem into computing similarity coefficients between cycles using signal processing techniques. This changes the parameter from a threshold-based metric to a similarity-based metric that does not require arbitrary threshold selection.
3Measurement precision
If standard deviation of cycle duration is used to assess regularity, then overall gait regularity can be measured, but the timing and isolation of erratic cycles are lost
Solution Approach 1:
The patent segments the gait analysis into individual cycle comparisons rather than computing a single aggregate statistic. Each cell in the similarity matrix represents a specific cycle comparison, preserving information about which cycles differ and when they occur in the sequence.
4Ease of manufacture
If parameters average gait characteristics over the entire exercise, then comprehensive gait assessment is achieved, but progress during the exercise is obscured
Solution Approach 1:
The patent introduces a matrix dimension where rows and columns represent different time points (cycles) in the exercise. This allows the analysis to simultaneously provide overall assessment through matrix patterns and temporal progression information through the arrangement of similarity coefficients across cycles.
Data Source
Figure 1A~2
Figure 3
Figure 4A~4B
AI summary
The present invention relates to a device 1 for analysing the regularity and symmetry of a sequence of N walking or running cycles of an individual, comprising sensors 2 for measuring raw time signals of a physical displacement variable of an anatomical segment, a processing unit connected to the measurement sensors 2 and having calculation and processing means arranged for separating the raw time signals into distinct time signals Ci, wherein the series Ci is associated with a given walking or running cycle i of the individual, calculating at least one coefficient of similarity between the signal Ci associated with the walking or running cycle i and another signal Cj associated with a walking or running cycle j of the same individual, display means 4 connected to the processing unit and displaying the matrix M(i,j) in which each value of the coefficient of similarity is represented by a colour located in a graduated colour scale to allow the naked eye to see the similarity between the walking or running cycles i and j.