Detection Of Dyslexia From Eye Movements Using Anfis & Bbwpe Feature Extraction Methods

ABSTRACT - The control of attention orienting was studied in children with specific reading disorder (SRD)<br>or dyslexia, and it was compared with that of normal readers. The main goal of the study was to propose and<br>implement new feature extraction method for extracting the features efficient from the signals and it is classified<br>using the classifier from eye movement signal. Eye movements of 76 school children were measured using<br>videooculographic (VOG) technique during one reading and four nonreading tasks. Videooculographic (VOG)<br>is used to measure the Eye movement’s signals of children through single reading and four non reading tasks.<br>Time and frequency domain features were extracted and various feature selection methods were performed to<br>select subsets of significant features. The best basis-based wavelet packet entropy method (BBWPE) is proposed<br>in this research for extracting the features from the eye movement signal. The improved ANFIS based on PSO is<br>used as a classifier. The parameters used for finding accuracy of the proposed methodology are sensitivity,<br>specificity and p-value. The Experimentation confirmed that the proposed ANFIS with BBWPE model has good<br>detection results than ANFIS with MAR and ANN model in terms of parameters like sensitivity, specificity,<br>detection accuracy and p-value.<br>KEYWORD - Learning disability, Feature Extraction, Classification, Adaptive Neuro-Fuzzy Inference System<br>(ANFIS), Best Basis-based Wavelet Packet Entropy method (BBWPE)