Applied Machine Learning on Phase of Gait Classification and Joint-Moment Regression

Erik Jung,Cheryl Lin,Mircea Teodorescu,Martin Contreras

doi:10.3390/biomechanics2010006

Erik Jung, Cheryl Lin + Show 2 more

Open Access

https://doi.org/10.3390/biomechanics2010006

Copy DOI

Abstract

Traditionally, monitoring biomechanics parameters requires a significant amount of sensors to track exercises such as gait. Both research and clinical studies have relied on intricate motion capture studios to yield precise measurements of movement. We propose a method that captures motion independently of optical hardware with the specific goal of identifying the phases of gait using joint angle measurement approaches like IMU (inertial measurement units) sensors. We are proposing a machine learning approach to progressively reduce the feature number (joint angles) required to classify the phases of gait without a significant drop in accuracy. We found that reducing the feature number from six (every joint used) to three reduces the mean classification accuracy by only 4.04%, while reducing the feature number from three to two drops mean classification accuracy by 7.46%. We extended gait phase classification by using the biomechanics simulation package, OpenSim, to generalize a set of required maximum joint moments to transition between phases. We believe this method could be used for applications other than monitoring the phases of gait with direct application to medical and assistive technology fields.

Highlights

Biomechanics is the study of human movement that combines the laws of physics with concepts of engineering to address physical health and performance [1]
Where the joint angle feature combinations are iterated through to find the ideal combination of the least number of features n and highest accuracy
Due to the nature of unique gait motions across individuals, we focused on the strong connection between phase of gait, joint angles, and joints’ moments

Summary

Introduction

Human gait produces a locomotion using the combination of the brain, nerves, and muscles in the lower extremities [2]. Balance and gait work uniformly as a complex sensory and motor coordination. Within this context, the assessment of gait indicates levels of physical mobility and effects of therapy or assistive technologies. A gait cycle starts at the point of initial contact of one lower extremity to the point where the same extremity touches the ground again [2]. Skeletal-based arrangements like the human body rely on muscles and tendons to manipulate joints [3]. The human leg depends on three primary joints: hip, knee, and ankle [4]

Objectives

Methods

Results

Conclusion