Gait Evaluation Using Procrustes and Euclidean Distance Matrix Analysis

Abstract

Objective assessment of gait is important in the treatment and rehabilitation of patients with different diseases. In this paper, we propose a gait evaluation system using Procrustes and Euclidean distance matrix analysis. We design and develop an android app to collect real time synchronous accelerometer and gyroscope data from two Inertial Measurement Unit (IMU) sensors through Bluetooth connectivity. The data is collected from 12 young (10 for modelling and 2 for validation) and 20 older subjects. We analyse the data collected from real world for stride, step, stance and swing gait features. We validate our method with measurements of gait features. Generalized Procrustes analysis is used to estimate a standard normal mean gait shape (NMGS) for 10 young subjects. Each gait feature of both young and older subjects is then converted to find the best match with the NMGS using ordinary Procrustes analysis. The shape distance between the NMGS and each gait shape is estimated using Riemannian shape distance, Riemannian size-and-shape distance, Procrustes size-and-shape distance and Root mean square deviation. A t-test is performed to provide statistical evidence of gait shape differences between young and older gaits. A mean form which is considered as a standard normal mean gait form (NMGF) and inter-feature distances are estimated from the set of 10 young subjects. The form difference is estimated between the NMGF and individual gaits of young and older. The degree of abnormality is then estimated for individual features and the result is plotted to visualize the feature in a gait. Experimental results demonstrate the performance of the proposed method. IndexTerms-Gait Analysis; Gait Assessment; Gait Features; Inertial Measurement Unit (IMU) I. INTRODUCTION UMAN gait is the result of a series of rhythmic alternating movements of the arms, legs, and trunk which create forward movement of the body [1]. Its complex mechanisms depend upon the integrated actions of the musculoskeletal, nervous system, visual, vestibular, auditory systems leading to the smooth propulsive movement of the centre of gravity. Quantification of gait variabilities, kinematic and kinetic measurements, muscular measurements and energy expenditure, provide comprehensive locomotive gait information [2]. Gait quantification information is used to 1) distinguish the type of gait impairments and suggest possible

Upper extremity kinematics: development of a quantitative measure of impairment severity and dissimilarity after stroke

“…Procrustes analysis is another such psychometric method of quantifying difference or dissimilarity between two sets of data ( Kendall, 1989 ). Procrustes distance has recently garnered attention as a metric in both gait ( Rida, Almaadeed & Almaadeed, 2019 ; Sehairi, Chouireb & Meunier, 2018 ; Anwary, Yu & Vassallo, 2019 ) and upper extremity studies ( Passos et al, 2023 ; Saenen, Orban de Xivry & Verheyden, 2022 ; Wong et al, 2019 ; Passos, Campos & Diniz, 2017 ). Procrustes analysis quantifies the similarity of shape between two matrix sets and provides the linear transformation that would allow one curve to best conform to the other.…”

Daily Living Activity Recognition In-The-Wild: Modeling and Inferring Activity-Aware Human Contexts

“…They used the conditional random field (CRF) classifier for AR task and acquired the best average accuracy around 95%. Anwary et al [57] utilized wearable sensors (i.e., accelerometer and gyroscope) for monitoring and detecting abnormalities in the gait pattern of the participants. Moreover, wearable sensors have also been utilized for detecting and preventing abrupt human actions, such as falls [20].…”

Context-Aware Human Activity Recognition (CAHAR) in-the-Wild Using Smartphone Accelerometer

“…The authors extracted timedomain features to train the model using Random Forest and Naïve Bayes classifier and achieved 92% average accuracy for Random Forest classifier. Anwary et al [15] presented a gait evaluation system that utilizes Procrustes and Euclidean distance matrix analysis to find out the degree of abnormality in the gait pattern of a person using the wearable accelerometer and gyroscope sensors. The authors collected data from twelve (12) and twenty (20) young and older subjects, respectively, and extracted gait features such as step, swing, stance, and stride for analyzing the gait pattern for any abnormality.…”

“…For standing activity, indoor and outdoor contexts are used. By incorporating selected contexts with the primary activities, fifteen (15) different context-aware activities are obtained, which are opted for CAHAR using Random Forest, Bagging, Decision Tree, K-Nearest Neighbors, Support Vector Machine, and Naïve Bayes classifier. For signifying the effectiveness of using a smartphone accelerometer for the proposed scheme, this paper also provides a comparison of the recognition performance of different sensors (phone accelerometer, phone gyroscope, and watch accelerometer) and their fusion for CAHAR.…”