Presentation 2006-01-28
Classification of ataxia using a self-organazing neural network model
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In clinical examination of ataxia, method of the quantitative analysis has been least established, and thus we have previously developed a computerized system for measuring and evaluating the upper-limb movement function. Based on parameter values measured by using this system, we could find the statistically significant differences between normal subjects and the patients suffering from the Parkinson's disease (PD) or the spinocerebeller degeneration (SCD), and subsequently, we developed a hybrid artificial neural network (HNN) model [consisting of a self-organizing map (SOM) followed by a backpropagation neural network (BNN)] which can automatically discriminate between the normal subjects and either the PD or SCD patients. In the present study, a two-dimensional SOM and a three-dimensional SOM was constructed by the learning simultaneously the three parameter sets from normal subjects, PD patients, SCD patients. The structures of those SOMs suggested a hightly variable nature of ataxia even within PD or SCD patients.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Parkinson's disease / Spino-cerebellar degeneration / movement functional disorder of the upper limbs / Hybrid Neural Network / quantity evaluation
Paper # MBE2005-120
Date of Issue

Conference Information
Committee MBE
Conference Date 2006/1/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

Paper Information
Registration To ME and Bio Cybernetics (MBE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Classification of ataxia using a self-organazing neural network model
Sub Title (in English)
Keyword(1) Parkinson's disease
Keyword(2) Spino-cerebellar degeneration
Keyword(3) movement functional disorder of the upper limbs
Keyword(4) Hybrid Neural Network
Keyword(5) quantity evaluation
1st Author's Name Tsuyoshi YAMAGUCHI
1st Author's Affiliation Kumamoto University Graduate School()
2nd Author's Name Tomohiko IGASAKI
2nd Author's Affiliation Kumamoto University
3rd Author's Name Yuuki HAYASHIDA
3rd Author's Affiliation Kumamoto University
4th Author's Name Nobuki MURAYAMA
4th Author's Affiliation Kumamoto University Graduate School:Kumamoto University
5th Author's Name Kikuo YAMAGUCHI
5th Author's Affiliation Kumamoto Saishunsou Hospital
Date 2006-01-28
Paper # MBE2005-120
Volume (vol) vol.105
Number (no) 578
Page pp.pp.-
#Pages 4
Date of Issue