Methodology proposal of ADHD classification of children based on cross recurrence plots

Dealing with electroencephalogram signals (EEG) is often not easy. The lack of predicability and complexity such non-stationary, noisy high-dimensional challenging. Cross recurrence plots (CRP) have been used extensively to deal the detection subtle changes in signals, even when noise embedded signal. In this contribution, a total 121 children performed visual attention experiments proposed methodology using CRP Welch Power Spectral Distribution classify then between those who ADHD control group. Additional tools were presented determine which extent able accurately avoid misclassifications, thus demonstrating that feasible EEG from subjects ADHD. experimental results indicate shows higher accuracy comparison methods by other authors, providing correct are selected. Also, does require extensive training as classical machine learning algorithms. Furthermore, it required manually discriminate events among electrodes since can detect smallest signal has noise. Lastly, compared baseline prove experimentally consistent repeatable. Given right metrics, an up 97.25% was achieved, indicating outperformed many state-of-the-art techniques.