Foot-Motion Gait Determination Using Relative Change Values

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing gait analysis systems struggle to accurately distinguish between normal and exceptional gaits on uneven terrain, leading to potential misclassification of health conditions, particularly for individuals with reduced muscle strength such as the elderly or rehabilitation patients.

Innovation Solution

A determination device that calculates a relative change value from a traveling axis for each gait cycle using time-series sensor data, employing a machine learning model to determine gait situations, including normal and exceptional gaits, and outputs gait information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If threshold-based filtering is used to remove exceptional gait data, then measurement precision is improved for normal subjects, but reliability deteriorates for subjects with reduced muscle strength

Engineering Contradiction:

Improvegait classification accuracyVSAvoidhealth condition assessment reliability

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent transforms the gait analysis approach by changing from simple threshold-based filtering to analyzing relative change values (derivatives) of sensor data. This parameter transformation enables differentiation between exceptional gaits caused by terrain and those caused by health issues, even for subjects with reduced muscle strength who cannot generate strong signal variations.

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If simple threshold values are applied to sensor data, then ease of operation is improved, but measurement precision deteriorates for distinguishing exceptional gait types

Engineering Contradiction:

Improvedata processing simplicityVSAvoidgait situation discrimination accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces relative change values as an intermediary parameter between raw sensor data and gait classification. By computing derivatives of acceleration and angular velocity data, the system creates a new representation that enhances the distinguishability of different gait situations while maintaining automated processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If data from exceptional gait situations is included in health assessment, then productivity is improved by using all available data, but measurement precision deteriorates due to misclassification

Engineering Contradiction:

Improvedata utilization efficiencyVSAvoidhealth condition determination accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments gait data into different situations (exceptional gait vs. normal gait) by analyzing relative change values. This segmentation allows the system to identify and exclude data from exceptional situations such as meandering roads or stairs, ensuring that only relevant data is used for health condition assessment while maintaining high data utilization efficiency.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20250295330A1Determination device, determination method, and recording

Publication Date: 2025.09.25 NEC CORP

AI summary

Provided is a determination device including a data acquisition unit that acquires sensor data measured in accordance with a motion of a foot, a calculation unit that calculates a relative change value indicating a relative change from a traveling axis for each gait cycle using time-series data of the acquired sensor data, a determination unit that determines a gait situation using time-series data of the relative change value in a target period, and an output unit that outputs gait information including the determined gait situation.