This project focuses on the development of a machine learning (ML) classifier designed to detect and compare major psychiatric disorders using electroencephalography (EEG) data. The dataset consists of 945 subjects, including 850 patients with major psychiatric disorders (covering six large-categorical and nine specific disorders) and 95 healthy controls (HCs). The data includes intelligence quotient (IQ) scores, psychological assessment results, and quantitative EEG (QEEG) parameters recorded during resting-state assessments. By analyzing power spectrum density (PSD) and functional connectivity (FC) across different frequency bands, we developed ML models to classify and differentiate psychiatric disorders from healthy controls. Various algorithms, including Support Vector Machine (SVM), Random Forest, and XGBoost, were utilized to construct binary classification models for each disorder. The output of the system provides classifications for conditions such as mood disorders, addictive disorders, trauma and stress-related disorders, schizophrenia, and anxiety disorders. The technologies employed in this project include Python, Django, Scikit-learn, Imblearn, and SQLite for efficient data processing and model development.

Keywords: Psychiatric Disorders, Electroencephalography (EEG), Quantitative EEG (QEEG), Machine Learning, Power Spectrum Density (PSD), Functional Connectivity (FC), Support Vector Machine (SVM), Random Forest, XGBoost, Mental Health Classification.