LSGP-USFNet: Automated Attention Deficit Hyperactivity Disorder Detection Using Locations of Sophie Germain’s Primes on Ulam’s Spiral-Based Features with Electroencephalogram Signals

Anxiety, learning disabilities, and depression are the symptoms of attention deficit hyperactivity disorder (ADHD), an isogenous pattern hyperactivity, impulsivity, inattention. For early diagnosis ADHD, electroencephalogram (EEG) signals widely used. However, direct analysis EEG is highly challenging as it time-consuming, nonlinear, nonstationary in nature. Thus, this paper, a novel approach (LSGP-USFNet) developed based on patterns obtained from Ulam’s spiral Sophia Germain’s prime numbers. The initially filtered to remove noise segmented with non-overlapping sliding window length 512 samples. Then, time–frequency approach, namely continuous wavelet transform, applied each channel signal interpret time frequency domain. representation saved image, n × image for patch extraction. An localized patch, gray levels acquired features where Sophie primes located spiral. All tones all patches concatenated construct ADHD normal classes. A tone selection algorithm, ReliefF, employed representative acquire final most important tones. support vector machine classifier used 10-fold cross-validation criteria. Our proposed LSGP-USFNet, was using publicly available dataset accuracy 97.46% detecting automatically. generated model ready be validated bigger database can also detect other children’s neurological disorders.