Insole-Based Estimation of Vertical Ground Reaction Force Using One-Step Learning with Probabilistic Regression and Data Augmentation

Ryo Eguchi, Masaki Takahashi

Research output: Contribution to journalArticle

Abstract

An insole-based estimation of the vertical ground reaction force (vGRF) is proposed as an alternative to costly force plates for the evaluation of pathological gait. However, machine learning techniques for estimation still rely on the use of force plates. Moreover, measuring plural walking steps in order to prevent overfitting induces fall risks and physically taxes the patients. Therefore, this paper presents an accessible and efficient learning scheme for the insole-based estimation of vGRF. In this system, we employ a low-cost scale as an alternative to force plates. Then, we use Gaussian process regression (GPR) to learn a model in order to estimate vGRF without overfitting of small-sized data sets corrupted by measurement errors and noise of the devices. In addition, we propose a 'one-step learning' scheme based on a probabilistic data augmentation. This approach augments actual measurements of a minimum (just one) walking step to a virtual data set for plural steps by considering their typical variability between steps. In experiments, the GPR models learned from two walking steps estimated vGRF with mean errors of 8% or under for entire/local magnitudes. Moreover, the learning from one step with probabilistic augmentation enhanced the estimation accuracy.

Original languageEnglish
Article number8713540
Pages (from-to)1217-1225
Number of pages9
JournalIEEE Transactions on Neural Systems and Rehabilitation Engineering
Volume27
Issue number6
DOIs
Publication statusPublished - 2019 Jun 1

Fingerprint

Walking
Learning
Taxes
Gait
Noise
Taxation
Measurement errors
Learning systems
Costs and Cost Analysis
Equipment and Supplies
Costs
Datasets
Experiments
Machine Learning

Keywords

  • data augmentation
  • estimation
  • Gait analysis
  • Gaussian process regression
  • ground reaction force
  • instrumented insole
  • probabilistic machine learning
  • Wii Balance Board

ASJC Scopus subject areas

  • Neuroscience(all)
  • Biomedical Engineering
  • Computer Science Applications

Cite this

@article{43452adcfde94ec3b102fdfad7504623,
title = "Insole-Based Estimation of Vertical Ground Reaction Force Using One-Step Learning with Probabilistic Regression and Data Augmentation",
abstract = "An insole-based estimation of the vertical ground reaction force (vGRF) is proposed as an alternative to costly force plates for the evaluation of pathological gait. However, machine learning techniques for estimation still rely on the use of force plates. Moreover, measuring plural walking steps in order to prevent overfitting induces fall risks and physically taxes the patients. Therefore, this paper presents an accessible and efficient learning scheme for the insole-based estimation of vGRF. In this system, we employ a low-cost scale as an alternative to force plates. Then, we use Gaussian process regression (GPR) to learn a model in order to estimate vGRF without overfitting of small-sized data sets corrupted by measurement errors and noise of the devices. In addition, we propose a 'one-step learning' scheme based on a probabilistic data augmentation. This approach augments actual measurements of a minimum (just one) walking step to a virtual data set for plural steps by considering their typical variability between steps. In experiments, the GPR models learned from two walking steps estimated vGRF with mean errors of 8{\%} or under for entire/local magnitudes. Moreover, the learning from one step with probabilistic augmentation enhanced the estimation accuracy.",
keywords = "data augmentation, estimation, Gait analysis, Gaussian process regression, ground reaction force, instrumented insole, probabilistic machine learning, Wii Balance Board",
author = "Ryo Eguchi and Masaki Takahashi",
year = "2019",
month = "6",
day = "1",
doi = "10.1109/TNSRE.2019.2916476",
language = "English",
volume = "27",
pages = "1217--1225",
journal = "IEEE Transactions on Neural Systems and Rehabilitation Engineering",
issn = "1534-4320",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

TY - JOUR

T1 - Insole-Based Estimation of Vertical Ground Reaction Force Using One-Step Learning with Probabilistic Regression and Data Augmentation

AU - Eguchi, Ryo

AU - Takahashi, Masaki

PY - 2019/6/1

Y1 - 2019/6/1

N2 - An insole-based estimation of the vertical ground reaction force (vGRF) is proposed as an alternative to costly force plates for the evaluation of pathological gait. However, machine learning techniques for estimation still rely on the use of force plates. Moreover, measuring plural walking steps in order to prevent overfitting induces fall risks and physically taxes the patients. Therefore, this paper presents an accessible and efficient learning scheme for the insole-based estimation of vGRF. In this system, we employ a low-cost scale as an alternative to force plates. Then, we use Gaussian process regression (GPR) to learn a model in order to estimate vGRF without overfitting of small-sized data sets corrupted by measurement errors and noise of the devices. In addition, we propose a 'one-step learning' scheme based on a probabilistic data augmentation. This approach augments actual measurements of a minimum (just one) walking step to a virtual data set for plural steps by considering their typical variability between steps. In experiments, the GPR models learned from two walking steps estimated vGRF with mean errors of 8% or under for entire/local magnitudes. Moreover, the learning from one step with probabilistic augmentation enhanced the estimation accuracy.

AB - An insole-based estimation of the vertical ground reaction force (vGRF) is proposed as an alternative to costly force plates for the evaluation of pathological gait. However, machine learning techniques for estimation still rely on the use of force plates. Moreover, measuring plural walking steps in order to prevent overfitting induces fall risks and physically taxes the patients. Therefore, this paper presents an accessible and efficient learning scheme for the insole-based estimation of vGRF. In this system, we employ a low-cost scale as an alternative to force plates. Then, we use Gaussian process regression (GPR) to learn a model in order to estimate vGRF without overfitting of small-sized data sets corrupted by measurement errors and noise of the devices. In addition, we propose a 'one-step learning' scheme based on a probabilistic data augmentation. This approach augments actual measurements of a minimum (just one) walking step to a virtual data set for plural steps by considering their typical variability between steps. In experiments, the GPR models learned from two walking steps estimated vGRF with mean errors of 8% or under for entire/local magnitudes. Moreover, the learning from one step with probabilistic augmentation enhanced the estimation accuracy.

KW - data augmentation

KW - estimation

KW - Gait analysis

KW - Gaussian process regression

KW - ground reaction force

KW - instrumented insole

KW - probabilistic machine learning

KW - Wii Balance Board

UR - http://www.scopus.com/inward/record.url?scp=85067464133&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067464133&partnerID=8YFLogxK

U2 - 10.1109/TNSRE.2019.2916476

DO - 10.1109/TNSRE.2019.2916476

M3 - Article

VL - 27

SP - 1217

EP - 1225

JO - IEEE Transactions on Neural Systems and Rehabilitation Engineering

JF - IEEE Transactions on Neural Systems and Rehabilitation Engineering

SN - 1534-4320

IS - 6

M1 - 8713540

ER -