Diagnosis of Alzheimer’s Disease, Parkinson’s Disease, Frontotemporal Dementia, and Paranoid Schizophrenia via Complex Network Analysis of EEG Data

A novel diagnostic method that employs complex network analysis using electroencephalogram (EEG) data is presented, which achieves exceptional classification accuracy across a range of mental disorders. Our method demonstrates 96% accuracy for paranoid schizophrenia, 96% for frontotemporal dementia, 97% for Alzheimer’s disease, and 96% for Parkinson’s disease, highlighting the robustness and versatility of this approach. The method’s efficacy lies in its ability to handle various data sets, including diverse channel configurations such as the 10-20 extended system for Parkinson’s disease, thereby ensuring broad applicability. Despite utilizing different data formats and sizes, the approach consistently achieves high precision. The method’s simplicity, computational efficiency, and scalability offer a significant advancement in neurodiagnostic applications.