Dyslexia Detection using Event-related potential and Ensemble learning

Document Type : Original Article

Authors

1 Biomedical group, Faculty of Electrical Engineering, K. N. Toosi University of Technology, Tehran, Iran,

2 - Faculty of Electrical Engineering, K.N.Toosi University of Technology, Tehran, Iran

3 Faculty of Education and Psychology, Shahid Beheshti University, Tehran, Iran

4 Faculty of Psychology and Education, University of Tehran, Tehran, Iran

Abstract

Dyslexia is a common learning disorder that is a combination of disabilities and problems that affect the learning process in one or more areas, including reading, writing, and spelling. The diagnosis of dyslexia is traditionally made using various diagnostic tests. In addition to requiring reading and writing skills, these tests are time-consuming and error-prone. For this reason, researchers have proposed a wide range of methods for dyslexia detection, including game-based techniques, recording and writing images, eye movement tracking, and functional and structural magnetic resonance imaging of the brain. This article aims to diagnose dyslexia using evoked-related potentials while performing a visual continuous performance task. For this purpose, we extracted Event-related potential (ERP) signal and used the principal component analysis method to reduce the number of dimensions in the data. Then the ensemble learning classification method was used to discriminate the data into two groups: normal group and dyslexic group. The proposed method performs well in the second event with 87/5% accuracy and 81/2% sensitivity in dyslexia detection. The obtained results demonstrate that with dyslexic children may be unable to control their initial decisions and assess things rapidly.

Keywords

Main Subjects

Advanced Signal Processing

Articles in Press, Accepted Manuscript
Available Online from 12 August 2022

Files

Share

How to cite

Statistics