An Adaptive Optimized Schizophrenia Electroencephalogram Disease Prediction Framework

Electroencephalogram (EEG) signal analysis has become an interesting and required area in the medical industry to analyze brain function for different diseases. But, the EEG signal’s noise features might degrade the signal prediction's exactness score. So, the presented article aims to develop a novel EEG signal analysis system named a novel Firefly-based Deep Belief Signal Specification (FbDBSS). In addition, the disease signal considered in this research work is Schizophrenia (SZ) signal. Initially, the SZ signal with a normal EEG signal is trained to the system, and preprocessing function is performed. Then the filtered signal is entered into the classification layer for the feature extraction and signal analysis function. Furthermore, the proposed design is executed in the python environment, and the robustness score has been measured in terms of accuracy, sensitivity, and error rate. The chief parameter of the proposed FbDBSS design is compared with other models and has gained the finest 3% of improved signal analysis accuracy and sensitivity score.