Gait Waveform Interpolation for Missing Data Continuity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for interpolating missing data in gait analysis struggle to accurately restore the characteristics of missing sections, especially in complex waveforms, and require pre-stored time-series models and interpolation methods.

Innovation Solution

An interpolation device that generates a gait waveform for each gait cycle using sensor data, specifies missing sections, calculates the gait phase of the missing sections, and interpolates these sections using data from gait waveforms with different cycles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing interpolation methods are used to restore missing data sections, then data continuity is improved, but the characteristics and features of the missing sections are lost

Engineering Contradiction:

Improvedata continuityVSAvoidfeature preservation

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The gait waveform is segmented into multiple phases (stance phase, swing phase, etc.) based on gait events. When interpolating missing data, the method identifies which phase the missing section belongs to and applies phase-specific interpolation strategies, preserving the characteristic features of each gait phase while restoring data continuity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The method transforms the interpolation approach by changing from generic time-series interpolation to gait-phase-specific interpolation. By parameterizing the interpolation process according to gait phase characteristics (such as using acceleration waveform features to identify stance/swing phases), the method preserves essential gait features while restoring missing data.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If pre-stored time-series models and interpolation methods are used, then interpolation capability is provided, but device complexity and storage requirements increase

Engineering Contradiction:

Improveinterpolation capabilityVSAvoidmodel storage requirement

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs self-adaptation by automatically identifying gait phases and selecting appropriate interpolation methods based on the actual data characteristics. Instead of requiring pre-stored models for various scenarios, the system analyzes the input data itself (using gait events detected from the data) to determine the appropriate interpolation strategy, eliminating the need for extensive pre-stored models.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The interpolation method dynamically adapts to different gait phases and missing data patterns. Rather than using a static pre-stored model, the system adjusts the interpolation approach based on real-time analysis of gait phase identification, making the system versatile without requiring large storage capacity for pre-computed models.

Inventive Principle:

Principle #15Dynamics

3Loss of information

If generic signal interpolation methods are applied, then missing data is restored, but gait-specific characteristics are not preserved

Engineering Contradiction:

Improvedata restorationVSAvoidgait feature accuracy

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The method applies different interpolation qualities and strategies to different portions of the gait waveform based on local characteristics. For example, during stance phase interpolation, the method preserves characteristics typical of ground contact, while during swing phase, it preserves characteristics of foot movement through air. This local quality approach ensures gait-specific features are maintained in each section.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

Before performing interpolation, the system performs preliminary analysis to identify gait phases and characteristic points (heel strike, toe-off, etc.). This preliminary action of phase identification enables the subsequent interpolation to target-specifically preserve relevant gait features rather than applying a generic restoration method.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250134412A1Interpolation device, gait measurement system, interpolation method, and recording medium

Publication Date: 2025.05.01 NEC CORP

AI summary

An interpolation device includes a gait information processing unit that generates a gait waveform for each gait cycle by using time-series data of sensor data regarding a movement of a foot and specifies a missing section of data in the time-series data, a missing information processing unit that calculates a gait phase of the specified missing section, and an interpolation unit that generates interpolation data for interpolating the missing section by using data of the gait phase of the missing section in a gait waveform with a gait cycle different from a gait cycle including the missing section, and interpolates the generated interpolation data to the missing section.