Harnessing the Potential of Light Gradient Boosting Machine for Accurate Diagnosis of Schizophrenia from EEG Signals

Schizophrenia is a psychiatric condition that distresses many individuals, leading to disturbances in cognitive faculties, emotional states, and interpersonal dynamics. Diagnosis using AI is imperative in healthcare, aiding patient care, intelligent health systems, risk identification of patients and drug discovery through various medical data sources like ultrasound, EEG, EOG, MRI and CT scans. Although many researchers have contributed to diagnosis and classification of schizophrenia. In this paper, we study the potential of Light Gradient Boosting Machine (Light GBM) algorithm in effective identification of schizophrenia using EEG signal data. Light GBM successfully captures intricate connections within the data. Our approach can handle large size datasets by dimensionality reduction and findings indicate that Light GBM demonstrates relatively better accuracy in detecting schizophrenia compared to alternative models proposed by different researchers with an exceptional accuracy of 98%. The model's resilience and capacity to apply knowledge to new data emphasize its clinical usefulness.