Loading...

EEG based Analysis and Classification of Children with Learning Disability Compared to Normal Children

Mirmohammad Sadeghi, Delaram Alsadat | 2017

655 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: English
  3. Document No: 49626 (55)
  4. University: Sharif University of Technology, International Campus, Kish Island
  5. Department: Science and Engineering
  6. Advisor(s): Jahed, Mehran
  7. Abstract:
  8. Learning disability (LD) is a neurological condition that interferes with an individual’s ability to store, process, or produce information. There are different types of learning disabilities affecting reading, writing, speaking, spelling, etc. Based on a study conducted by National Center for Learning Disabilities, 2.4 million American public school students are diagnosed with learning disability. They attend school in order to learn and be successful while they do not know their learning process is different from their peers. LD diagnosis in children is especially important as such cases must be identified early enough in order to provide them with proper education.This project targets LD children on the basis that their brains may function differently from normal children. The first goal of this research is to analyze electroencephalogram (EEG) of children diagnosed with LD to evaluate possible differences in their brain activity compared to normal children. The second goal is to implement a classification algorithm to identify LD children among normal children with high accuracy. To do so, EEG signals from 12 children diagnosed with LD and 3 normal children were recorded.Wavelet analysis was used to extract features from EEG signals followed by Support Vector Machine (SVM) as a classification algorithm. Comparing the frequency band powers of EEG in LD and normal children revealed that except for the delta band all other EEG frequency bands have higher activity in LD children. Proposed EEG signal classification using SVM achieved 83% accuracy while the feature selection algorithm improved the accuracy up to 96.7%
  9. Keywords:
  10. Electroencephalography ; Classification ; Wavelet Analysis ; Support Vector Machine (SVM) ; Learning Disability

 Digital Object List

 Bookmark

No TOC