Detection of Attention Deficit Hyperactivity Disorder from EEG Signal using Discrete Wavelet Transform

Attention deficit hyperactivity disorder (ADHD) is the most prevalent heterogeneous neurodevelopment disorder in children and its diagnosis is based on symptom questionnaires, neuropsychological testing and clinical interviews. There is a great need for early identification of children with ADHD so that remedial action can be initiated in very early stage itself that in turn will improve the academic and behavioral performances. The crux of this study is to identify ADHD from EEG signals. In this study, 10 samples participated including 5 with ADHD and 5 with the control group. The electroencephalogram signals are recorded by 16 electrodes in eyes-open and eyes-closed resting conditions. Feature extraction is analyzed by using a Discrete Wavelet Transform. The feature extracted signals are classified using Support Vector Machine and the classification is analyzed. A significant difference has been observed between the samples that have ADHD and the normal subjects.