This research investigates the effectiveness of neurofeedback (NF) as a non-pharmacological intervention for children diagnosed with ADHD, combining EEG analysis with advanced machine learning techniques. EEG signals were collected and optimized using SVM-based feature selection and t-tests, identifying six highly significant channels linked to ADHD-related brain activity. A meta-analysis of existing clinical studies showed measurable improvements in attention control, reduced hyperactivity, and lower impulsivity following NF therapy. Multiple ML models—including Random Forest, k-NN, Decision Tree, and Logistic Regression—were evaluated, with the Gaussian Process (RBF kernel) achieving the best performance. This work highlights the potential of integrating neurofeedback with computational intelligence for more accurate ADHD assessment and personalized treatment. 

Paper Link: Click Here