Paper Info
Title
Software development of an intelligent Spirography test system for neurological disorder detection and quantification
Abstract
Heretofore several efforts have been made for detection and quantification of neurological disorders which have observable symptoms as hand tremor. Multiple sclerosis is among such disorders which can somewhat quantified by measuring the severity of hand tremor. In this paper, a system is designed for recording and analysis of digital signal of Spirography standard test for this purpose. Hardware and software development are described for an apparatus, its performance is to make the standard Spirography test, to record the signal, to transfer the signal to the PC in which the associated software is installed and to analyze the signal according to the feature extraction and classification algorithms. Power Spectrum Analysis is proposed as one of the extracted features in the software since it reveals the effect of each frequency components in overall movement of hand. In addition to Power Spectrum Analysis complex features as Largest Lyapunov Exponent and mean value of the Lyapunov spectrum of the signals which are chosen to be the indications of the signals chaoticity level. Signal complexity is represented as its embedding dimension and time lag which together construct an approximate index window in periodic signal reconstruction manner. Time lag correlates to the sampling rate and signal geometry. Signals are treated as patterns in features space and they are undergone classification by a trained feed forward neural network. Classification task acts as the decision making process in which the membership of each subjects signal to the predefined classes of healthy and unhealthy group is calculated and corresponding consequent treatments are arranged by the physicist. It is shown in this paper that the complex features as chaotic features can representatively exhibit the signals dynamical behavior and they can be used for signal discrimination of subjects with and without hand tremor.
Year
DOI
Venue
2015
10.3233/IFS-141496
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
Keywords
Field
DocType
System development,graphic tablet WACOM,complex features,hand tremor,spiral drawing signal,MATLAB,signal discrimination
Feedforward neural network,Digital signal,Sampling (signal processing),Feature extraction,Software,Spectral density,Artificial intelligence,Statistical classification,Mathematics,Lyapunov exponent,Machine learning
Journal
Volume
Issue
ISSN
28
5
1064-1246
Citations 
PageRank 
References 
0
0.34
4
Authors
3
Name
Order
Citations
PageRank
Hadi Chahkandi Nejad1153.35
Omid Khayat2556.52
Javad Razjouyan383.44