1. Automatic Diagnosis of Schizophrenia in EEG Signals Using CNN-LSTM Models
- Author
-
Afshin Shoeibi, Yinan Kong, Saeid Nahavandi, Ali Khadem, Jónathan Heras, Roohallah Alizadehsani, Juan Manuel Górriz, Parisa Moridian, Navid Ghassemi, Yudong Zhang, and Delaram Sadeghi
- Subjects
Signal Processing (eess.SP) ,FOS: Computer and information sciences ,Normalization (statistics) ,Computer Science - Machine Learning ,diagnosis ,Computer science ,Activation function ,Biomedical Engineering ,Neuroscience (miscellaneous) ,Decision tree ,Neurosciences. Biological psychiatry. Neuropsychiatry ,Neuroimaging ,Electroencephalography ,Machine Learning (cs.LG) ,Bayes' theorem ,Diagnosis ,FOS: Electrical engineering, electronic engineering, information engineering ,EEG signals ,medicine ,Electrical Engineering and Systems Science - Signal Processing ,Original Research ,neuroimaging ,medicine.diagnostic_test ,business.industry ,Deep learning ,deep learning ,Pattern recognition ,Computer Science Applications ,Random forest ,schizophrenia ,Support vector machine ,Schizophrenia ,Artificial intelligence ,business ,EEG Signals ,RC321-571 ,Neuroscience - Abstract
This work was supported by the MCIN/AEI/10. 13039/501100011033/and FEDER "Una manera de hacer Europa" under the RTI2018-098913-B100 project, by the Consejeria de Economia, Innovacion, Ciencia y Empleo (Junta de Andalucia) and FEDER under CV20-45250, A-TIC-080-UGR18, B-TIC-586-UGR20, and P20-00525 projects., 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., FEDER "Una manera de hacer Europa" RTI2018-098913-B100, Junta de Andalucia European Commission CV20-45250 A-TIC-080-UGR18 B-TIC-586-UGR20 P20-00525 MCIN/AEI/10. 13039/501100011033/
- Published
- 2021