A Markerless-based Gait Analysis and Visualization Approach for ASD Children

Abstract

This study proposed a new method in gait acquisition and analysis for autistic children based on the markerless technique versus the gold standard marker-based technique. Here, the gait acquisition stage is conducted using a depth camera with a customizable skeleton tracking function that is the Microsoft Kinect sensor for recording the walking gait trials of the 23 children with autism spectrum disorder (ASD) and 30 typically healthy developing (TD) children. Next, the Kinect depth sensor outputs information is translated into kinematic gait features. Further, analysis and evaluation are done specifically the kinematic angles of the hip, knee, and ankle in analyzing and visualizing the pattern of the plots versus the kinematic plots acquired from the marker-based that is the Vicon motion system gait technique. In addition, these kinematic angles are also validated using the statistical method namely the Analysis of Variance (ANOVA). Results showed that the ρ-values are insignificant for all angles upon computing both the intragroup and inter-group normalization. Hence, these findings have proven that the proposed markerless-based gait technique is indeed apt to be used as a new alternative markerless method for gait analysis of ASD children.

Classification of Autism and Control Gait in Children Using Multisegment Foot Kinematic Features

“…Similarly, the highest reported classification accuracy of 99.3% was achieved using principal component analysis and linear discriminant analysis for data reduction on full-body three-dimensional joint position data in children with ASD and controls [ 15 ]. A recent study extracted kinematic features from a markerless-based data acquisition [ 16 ] that resulted in a classification accuracy of 92% using a rough set classifier [ 19 ]. In contrast, previous research has investigated the classification of ASD and control gait patterns using temporal-spatial, kinematic, and kinetic features [ 7 , 14 , 15 , 16 , 17 , 18 ].…”

“…A recent study extracted kinematic features from a markerless-based data acquisition [ 16 ] that resulted in a classification accuracy of 92% using a rough set classifier [ 19 ]. In contrast, previous research has investigated the classification of ASD and control gait patterns using temporal-spatial, kinematic, and kinetic features [ 7 , 14 , 15 , 16 , 17 , 18 ]. High classification accuracy of 95.8% has been achieved by combining all of these features using a support vector machine (SVM) classifier [ 14 ].…”

“…To further understand gait patterns in children with ASD versus neurotypical controls, there has been a recent increase in the number of studies using machine learning (ML) tools for classification based on combinations of kinematic and kinetic features [ 7 , 14 , 15 , 16 , 17 , 18 , 19 ]. Such studies facilitate the identification of discriminatory gait features in ASD versus neurotypical controls, which may lead to the development of automated gait classifiers, optimal treatment plans, improved functionality for children, and reduced health care costs.…”

“…Previous research has investigated the classification of ASD and control gait patterns using temporal-spatial, kinematic, and kinetic features [ 7 , 14 , 15 , 16 , 17 , 18 , 19 ]. Ilias et al, (2006) observed that the fusion of all these features resulted in a classification accuracy of 95.80% (sensitivity 100%, specificity 85.00%) using a support vector machine (SVM) classifier [ 14 ].…”

“…Other dimensionality reduction techniques employed data transformations, such as principal component analysis and linear discriminant analysis [ 15 ] on three-dimensional joint position data, which resulted in an accuracy of 99.3% (sensitivity 99.66%, specificity 99.00%). More recently, a study extracted kinematic features from a markerless-based data acquisition [ 16 ] that resulted in a classification accuracy of 92% using a rough set classifier [ 19 ].…”