Ambiguity domain-based identification of altered gait pattern in ALS disorder

Abstract

The onset of a neurological disorder, such as amyotrophic lateral sclerosis (ALS), is so subtle that the symptoms are often overlooked, thereby ruling out the option of early detection of the abnormality. In the case of ALS, over 75% of the affected individuals often experience awkwardness when using their limbs, which alters their gait, i.e. stride and swing intervals. The aim of this work is to suitably represent the non-stationary characteristics of gait (fluctuations in stride and swing intervals) in order to facilitate discrimination between normal and ALS subjects. We define a simple-yet-representative feature vector space by exploiting the ambiguity domain (AD) to achieve efficient classification between healthy and pathological gait stride interval. The stride-to-stride fluctuations and the swing intervals of 16 healthy control and 13 ALS-affected subjects were analyzed. Three features that are representative of the gait signal characteristics were extracted from the AD-space and are fed to linear discriminant analysis and neural network classifiers, respectively. Overall, maximum accuracies of 89.2% (LDA) and 100% (NN) were obtained in classifying the ALS gait.

Gait pattern discrimination of ALS patients using classi cation methods

“…According to the table, it is seen that the method having the best performance for BC1 is the time frequency-based study proposed by Sugavaneswaran et al They obtained 100% accuracy for BC1. It is noticed that the accuracy evaluated using ATAT data in their study was 100% [13]. This is higher than 96.55% calculated using only testing data in our study.…”

“…The fundamental novelty of this study is applying the features of CFS to the classification methods to obtain the best classification algorithm for the discrimination of ALS. The data used in previous studies in the literature included 5-min recordings [13,16,19,20,23]. These types of recordings may cause difficulties and discontinuities during gait.…”

“…Signal turn-count-based [16] LS-SVM 89.66 Statistical analysis [19] LS-SVM 90.32 Time frequency-based [13] NN-based 100 Wavelet-based [20] …”

“…Sugavaneswaran et al achieved 89.2% linear and 100% nonlinear accuracy in classification between Co. and ALS using a time frequency-based algorithm [13]. Liao et al proposed a method based on multiresolution entropy analysis of stance time fluctuations in an investigation of gait asymmetry.…”

“…Xia et al achieved 92.86% accuracy between ALS and Co. using entropy-based analysis [23]. The studies in the literature thus clearly show that the dynamics of gait signals show differences among Co. and subjects having neurological disorders [12,13,[16][17][18][19][20][21][22][23][24][25][26][27][28]. These types of studies have generally focused on stride interval time series analyses.…”