Walking Pattern Identification Using Multi-Axis Stride Index Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques face challenges in identifying and extracting measured values during normal walking periods without supervision, particularly at home or outdoors, due to difficulties in distinguishing normal walking from other movements like acceleration, deceleration, or direction changes, which hinders the assessment of a patient's health condition.
Innovation Solution
An information processing program that acquires and analyzes time-series data from sensor devices to calculate index values for different stride sections, determining whether each section corresponds to normal walking by comparing statistical values across axial directions, allowing for the identification of normal walking sections even without supervision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If sensor data is collected during unsupervised periods at home or outdoors, then the duration and quantity of health assessment data is improved, but the difficulty of distinguishing normal walking from other movements increases
Solution Approach 1:
The patent segments the continuous sensor data into discrete stride sections based on detected swing phases. By dividing the walking period into individual strides and analyzing each stride's characteristics separately, the system can identify normal walking patterns more accurately amidst various movements. This segmentation allows for precise measurement of walking-specific parameters while filtering out non-walking activities.
Solution Approach 2:
The patent introduces a new dimension of analysis by calculating index values that represent relationships between measured values in different axial directions. Instead of relying on single-axis acceleration data, the system computes ratios of measured values across multiple axes (e.g., vertical vs. horizontal components), creating a multi-dimensional characterization of movement that distinguishes normal walking from other activities more reliably.
2Measurement precision
If index values are calculated by comparing measured values in different axial directions, then the precision of normal walking identification is improved, but the complexity of data processing increases
Solution Approach 1:
The patent performs preliminary detection of swing phases and segmentation of stride sections before calculating the index values. By pre-identifying walking periods and dividing them into discrete strides, the system reduces the complexity of subsequent index calculations. This preliminary action ensures that the complex multi-axis comparisons are only performed on relevant data segments, optimizing processing efficiency.
Solution Approach 2:
The system uses the sensor data itself to automatically identify swing phases and segment strides without requiring external supervision or manual intervention. The measured values from the sensor device serve dual purposes: both detecting the presence of walking and providing the data needed for index calculation, thereby simplifying the overall processing architecture despite the computational complexity.
Data Source
AI summary
An information processing program that causes a computer to execute a process includes acquiring first time-series data that includes measured values regarding movement of legs of a subject in a first period in which the subject travels in a walking manner for each of a plurality of axial directions; acquiring a first index value that relates to value of the measured values of one axial direction of the plurality of axial directions over value of the measured values of another axial direction of the plurality of axial directions for a first section in which the subject swings one of the legs for each of first stride sections that corresponds to one step of the subject in the first period based on the acquired first time-series data; acquiring second time-series data that includes second measured values regarding the movement of the legs of the subject in a second period different from the first period in which the subject travels in the walking manner for each of the plurality of axial directions; acquiring a second index value that relates to value of the measured values of one axial direction of the plurality of axial directions over value of the measured values of another axial direction of the plurality of axial directions for a second section in which the subject swings one of the legs for each of second stride sections that corresponds to one step of the subject in the second period based on the acquired second time-series data; and determining whether or not each of the second stride sections in the second period corresponds to the walking manner, based on the acquired first index value and the acquired second index value.