Automatic Diagnosis of Schizophrenia in EEG Signals Using CNN-LSTM Models

Abstract

Schizophrenia (SZ) is a mental disorder whereby due to the secretion of specific chemicals in the brain, the function of some brain regions is out of balance, leading to the lack of coordination between thoughts, actions, and emotions. This study provides various intelligent deep learning (DL)-based methods for automated SZ diagnosis via electroencephalography (EEG) signals. The obtained results are compared with those of conventional intelligent methods. To implement the proposed methods, the dataset of the Institute of Psychiatry and Neurology in Warsaw, Poland, has been used. First, EEG signals were divided into 25 s time frames and then were normalized by z-score or norm L2. In the classification step, two different approaches were considered for SZ diagnosis via EEG signals. In this step, the classification of EEG signals was first carried out by conventional machine learning methods, e.g., support vector machine, k-nearest neighbors, decision tree, naïve Bayes, random forest, extremely randomized trees, and bagging. Various proposed DL models, namely, long short-term memories (LSTMs), one-dimensional convolutional networks (1D-CNNs), and 1D-CNN-LSTMs, were used in the following. In this step, the DL models were implemented and compared with different activation functions. Among the proposed DL models, the CNN-LSTM architecture has had the best performance. In this architecture, the ReLU activation function with the z-score and L2-combined normalization was used. The proposed CNN-LSTM model has achieved an accuracy percentage of 99.25%, better than the results of most former studies in this field. It is worth mentioning that to perform all simulations, the k-fold cross-validation method with k = 5 has been used.

Schizophrenia Diagnosis via FFT and Wavelet Convolutional Neural Networks utilizing EEG signals

“…Shu Lih, et al [13] with 28 subjects, 14 in each group performed an eleven-layered convolutional neural network (CNN) model in order to classify and extract features for signals. Shoeibi et al [14] investigated 14 subjects in each schizophrenia and control group. They applied different conventional machine learning methods and deep learning architectures which CNN-LSTM combination achieved the most promising result.…”

Multi-Scale Attention Convolutional Network for Masson Stained Bile Duct Segmentation from Liver Pathology Images

“…AlexNet [ 5 ], a classical convolutional neural network proposed by Krizhevsky et al, reached a new achievement in the ImageNet Large Scale Visual Recognition Challenge (LSVRC) in 2012, and is therefore leading to a great interest in convolutional networks among researchers. Compared to earlier image processing methods, convolutional neural networks have been improved more effectively, and therefore be applied in a large number of different fields including image classification [ 5 , 6 ], semantic segmentation [ 7 ], object detection [ 8 ], object generation [ 9 ], human pose estimation [ 10 ], diagnosis of diseases using EEG signals [ 11 , 12 ], and also an amount of applications in medical field.…”

Diagnostics of Articular Cartilage Damage Based on Generated Acoustic Signals Using ANN—Part II: Patellofemoral Joint

“…The development of diagnostic tools providing an alternative to classical imaging diagnostic methods, enabling inexpensive, noninvasive, and precise diagnosis of osteoarthritis even at an early stage seems to be extremely important in clinical practice. Automatic diagnosis of diseases using machine learning methods is now widely used in medicine [ 25 , 26 , 27 , 28 , 29 ]. Machine learning is most commonly is utilized in radiology for object detection, segmentation, and classification [ 30 ] A growing interest can also be observed in nuclear medicine, where machine learning is utilized, e.g., for the glomerular filtration rate in SPECT/CT [ 31 ].…”