Cognitive Computation

, Volume 2, Issue 3, pp 165–169

Computer-Aided Diagnosis of Ataxia SCA-2 Using a Blind Source Separation Algorithm

Authors

  • Rodolfo V. García
    • Network DepartmentUniversity of Holguín
    • University of Granada
    • Department of Computer Architecture and Technology, ETS Ing. Informática y de TelecomunicaciónUniversity of Granada
  • Carlos G. Puntonet
    • Department of Computer Architecture and Technology, ETS Ing. Informática y de TelecomunicaciónUniversity of Granada
  • Belén San Román
    • University of Granada
  • Luis Velázquez
    • Centre for the Research and Rehabilitation of Hereditary Ataxias “Carlos J. Finlay”
  • Roberto Rodríguez
    • Centre for the Research and Rehabilitation of Hereditary Ataxias “Carlos J. Finlay”
Article

DOI: 10.1007/s12559-010-9049-0

Cite this article as:
García, R.V., Rojas, F., Puntonet, C.G. et al. Cogn Comput (2010) 2: 165. doi:10.1007/s12559-010-9049-0

Abstract

This work discusses a new approach for ataxia SCA-2 diagnosis based on the application of independent component analysis to the data obtained by electro-oculography in several experiments carried out over healthy and sick subjects. Abnormalities in the oculomotor system are well-known clinical symptoms in patients of several neurodegenerative diseases, including modifications in latency, peak velocity, and deviation in saccadic movements, causing changes in the waveform of the patient response. The changes in the morphology waveform suggest a higher degree of statistic independence in sick patients when compared to healthy individuals regarding the patient response to the visual saccadic stimulus modeled by means of digital generated saccade waveforms. The electro-oculogram records of thirteen patients diagnosed with ataxia SCA2 (a neurodegenerative hereditary disease) and thirteen healthy subjects used as control were processed to extract saccades.

Keywords

Independent component analysisBlind source separationBiomedical engineeringComputer-aided diagnosis

Copyright information

© Springer Science+Business Media, LLC 2010