An SVM based algorithm for analysis and discrimination of dyslexic readers from regular readers using ERPs

• Published in: 2012 IEEE 27th Convention of Electrical and Electronics Engineers in Israel pp. 1 - 4
• Main Authors: Frid, A., Breznitz, Z.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2012
• Subjects: Classification; Classification algorithms; Dyslexia Disability; Educational institutions; Electrodes; Event Related Potentials (ERP) Processing; Feature extraction; Information processing; Support Vector Machine (SVM); Support vector machines; Training
• ISBN: 9781467346825; 1467346829
• DOI: 10.1109/EEEI.2012.6377068

Dyslexia is a learning disability that impairs a person's ability to decode words accurately and fluently. This deficit can manifest itself in the language-related domain as difficulties in phonological and orthographic working memory, brain systems asynchrony, poor executive function skills and/or poor rapid naming processing. However it is not clear yet whether the dyslexia phenomenon is only related to language or if it can also be seen as a non-language deficit. Moreover, if it is also related to non-language activity, it is important to verify if it is possible to identify dyslexic readers at the earliest stage of information processing for better and effective remediation. Based on this, an effective algorithm was developed for analysis and classification of subjects as either Regular Readers or Dyslexic Readers, by using EEG recorded channels with Event Related Potentials (ERP) methodology during an auditory, short non-linguistic, simple, sub-phonetic choices reaction time task.