Schizophernia Detection with Multichannel EEG Using HMM Model

Schizophrenia is a severe neurological and physiological disorder that alters a patient’s perception of reality. Schizophrenia causes a variety of symptoms, including hallucinations, delusions, aberrant behaviour, and thinking. It influences their professional, academic, personal and social lives. The proposed method involves the detection of disease EEG using Empirical Mode Decomposition (EMD) with feature extraction. The proposed network with Hidden Network Model (HMM) for slower range of frequency of alpha and faster range frequency of gamma in EEG waves. A bidirectional long short term memory (Bi-LSTM) layer-based model for diagnosing schizophrenia. The decision tree makes the decision among the diseased frequency range with schizophrenia range. Ultimately, the accuracy, sensitivity, specificity of suggested system are used to examine the performance of the ensemble Bi-LSTM and Decision tree. The model demonstrates an accuracy of 93% highlighting its effectiveness in detecting schizophrenia from EEG signals.