Correlation Among Joint Motions Allows Classification of Parkinsonian Versus Normal 3-D Reaching

Abstract

In this paper, an objective assessment for determining whether a person has Parkinson disease is proposed. This is achieved by analyzing the correlation between joint movements, since Parkinsonian patients often have trouble coordinating different joints in a movement. Thus, the auto-correlation coefficient of single joint movements and the cross-correlation between movements in a pair of joints (hand, wrist, elbow and shoulder) were studied. These features were used to train and provide classification of subjects as having or not having Parkinson's disease using the Least Square Support Vector Machine. Experimental results showed that using either auto-correlation or cross-correlation features for classification provided over 91% correct classification. Using both features together provided better performance (96.0%) than using either feature alone. In addition, the performance of LS-SVM is better than that of Self Organizing Map (SOM) and k-nearest neighbor (KNN) in this case.

Differentiation between Parkinson disease and other forms of Parkinsonism using support vector machine analysis of susceptibility-weighted imaging (SWI): initial results

“…Note that the accuracy is calculated as average accuracy of ten repetitions using ten-fold cross validation (average and standard deviation) of PD. The majority of these studies applied SVM classifiers to behavioral data of gait analysis [29], fine motor force tracking [30], analysis of wearable accelerometer sensors [31] or joint movement [32], and even voice recording [33]. The only previous SVM application to MRI data in the domain of PD analysed VBM pre-processed grey matter in 21 PD, 11 MSA-P and 10 PSP, and 22 healthy controls [34].…”

The rates of change of the stochastic trajectories of acceleration variability are a good predictor of normal aging and of the stage of Parkinson's disease

“…The frequency distribution of the kinematics data determines the types of statistical analyses that can be validly carried out on the data (Limpert et al, 2001; Limpert and Stahel, 2011). The PD literature often uses ANOVA (analyses of variance) and regression analyses to test a hypothesis see for example (Poizner et al, 1998; Messier et al, 2007; Chan et al, 2010; Mera et al, 2011; Sande De Souza et al, 2011; Venkatakrishnan et al, 2011) just to cite a few. However, these methods require that the data be normally distributed.…”

Characterization of spatiotemporally complex gait patterns using cross-correlation signatures

“…For example, Pohl et al used the cross-correlation technique to examine the joint couplings between rearfoot and four adjacent joint variables during [ ( F i g . _ 1 ) T D $ F I G ] walking, and Chan et al showed that cross-correlation of upper extremity joint movements could be used for classification between normal and Parkinsonian individuals when reaching for a target [22,23]. In contrast with these past studies, in the present effort, we seek to characterize joint couplings throughout the entire locomotive apparatus using a single representative diagram.…”

“…For example, cross-correlation has been used to identify the synchronization of motor unit firings or to perform pattern recognition of EMG signals [14,18]. Cross-correlation between body movements has also been used [22,23]. For example, Pohl et al used the cross-correlation technique to examine the joint couplings between rearfoot and four adjacent joint variables during [ ( F i g .…”