EEG based Schizophrenia Detection using SPWVD-ViT Model

Year 2022, , 137 - 144, 30.12.2022

Mesut Şeker , Mehmet Siraç Özerdem

https://doi.org/10.36222/ejt.1192140

Cited By: 1

Abstract

Schizophrenia is a typical neurological disease that affects patients’ mental state, and daily behaviours. Combining image generation techniques with effective machine learning algorithms may accelerate treatment process, and possible early alert systems prevents diseases from reaching out crucial phase. The purpose of current study is to develop an automated EEG based schizophrenia detection with the Vision Transformer (ViT) model using Smoothed Pseudo Wigner Ville Distribution (SPWVD) time-frequency input images. EEG recordings from 35 schizophrenia (sch) and 35 healthy conditions (hc) are analyzed. We have used 5-fold cross validation for evaluation and testing of the method. Classification task is carried out as subject-independent and subject-dependent method. We reached out overall accuracy of 87% for subject-independent and 100% for subject-dependent approach for binary classification. While ViT has ben extensively used in Natural Language Processing (NLP) field, dividing input images within a sequence of embedded image patches via. transformer encoder is a practical way for medical image learning and developing diagnostic tools. SPWVD-ViT model is recommended as a disease detection tool not only for schizophrenia but other neurological symptoms.

Keywords

EEG, Schizophrenia, Neurological Disease, Vision Transformer, Diagnosis, Detection, Time-frequency image

References

• [1] V. Rajinikanth, S. C. Satapathy, S. L. Fernandes, and S. Nachiappan, “Entropy based segmentation of tumor from brain MR images – a study with teaching learning based optimization,” Pattern Recognit. Lett., vol. 94, pp. 87–95, 2017.
• [2] “Schizophrenia.” [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/schizophrenia. [Accessed: 10-Jan-2022].
• [3] Z. Wang and T. Oates, “Imaging time-series to improve classification and imputation,” IJCAI Int. Jt. Conf. Artif. Intell., vol. 2015-Janua, no. Ijcai, pp. 3939–3945, 2015.
• [4] M. Seker and M. S. Ozerdem, “EEG Coherence as a Neuro-marker for Diagnosis of Schizophrenia,” in 2020 28th Signal Processing and Communications Applications Conference, SIU 2020 - Proceedings, 2020.
• [5] J. W. Kim, Y. S. Lee, D. H. Han, K. J. Min, J. Lee, and K. Lee, “Diagnostic utility of quantitative EEG in un-medicated schizophrenia,” Neurosci. Lett., vol. 589, pp. 126–131, 2015.
• [6] Z. Dvey-Aharon, N. Fogelson, A. Peled, and N. Intrator, “Schizophrenia detection and classification by advanced analysis of EEG recordings using a single electrode approach,” PLoS One, vol. 10, no. 4, pp. 1–12, 2015.
• [7] J. K. Johannesen, J. Bi, R. Jiang, J. G. Kenney, and C.-M. A. Chen, “Machine learning identification of EEG features predicting working memory performance in schizophrenia and healthy adults,” Neuropsychiatr. Electrophysiol., vol. 2, no. 1, pp. 1–21, 2016.
• [8] V. Jahmunah et al., “Automated detection of schizophrenia using nonlinear signal processing methods,” Artif. Intell. Med., vol. 100, no. June, p. 101698, 2019.
• [9] S. L. Oh, J. Vicnesh, E. J. Ciaccio, R. Yuvaraj, and U. R. Acharya, “Deep convolutional neural network model for automated diagnosis of Schizophrenia using EEG signals,” Appl. Sci., vol. 9, no. 14, 2019.
• [10] C. R. Phang, F. Noman, H. Hussain, C. M. Ting, and H. Ombao, “A Multi-Domain Connectome Convolutional Neural Network for Identifying Schizophrenia from EEG Connectivity Patterns,” IEEE J. Biomed. Heal. Informatics, vol. 24, no. 5, pp. 1333–1343, 2020.
• [11] J. Tudela, M. Martínez, R. Valdivia, J. Romo, M. Portillo, and R. Rangel, “On the use of pairwise distance learning for brain Signal classification with limited observations,” Nature, vol. 388. pp. 539–547, 2010.
• [12] C. A. T. Naira and C. J. L. Del Alamo, “Classification of people who suffer schizophrenia and healthy people by EEG signals using deep learning,” Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 10, pp. 511–516, 2019.
• [13] S. S. Daud and R. Sudirman, “Butterworth Bandpass and Stationary Wavelet Transform Filter Comparison for Electroencephalography Signal,” Proc. - Int. Conf. Intell. Syst. Model. Simulation, ISMS, vol. 2015-Octob, pp. 123–126, 2015.
• [14] X. Jiang, G. Bin Bian, and Z. Tian, “Removal of artifacts from EEG signals: A review,” Sensors (Switzerland), vol. 19, no. 5, pp. 1–18, 2019.
• [15] Z. Aslan and M. Akin, “A deep learning approach in automated detection of schizophrenia using scalogram images of EEG signals,” Physical and Engineering Sciences in Medicine, vol. 45, no. 1. pp. 83–96, 2022.
• [16] Z. Aslan and M. Akin, “Automatic detection of schizophrenia by applying deep learning over spectrogram images of EEG signals,” Trait. du Signal, vol. 37, no. 2, pp. 235–244, 2020.
• [17] A. Zülfikar and A. Mehmet, “Empirical mode decomposition and convolutional neural network-based approach for diagnosing psychotic disorders from eeg signals,” Appl. Intell., vol. 52, no. 11, pp. 12103–12115, 2022.
• [18] K. Jindal, R. Upadhyay, P. K. Padhy, and L. Longo, “6 - Bi-LSTM-deep CNN for schizophrenia detection using MSST-spectral images of EEG signals,” in Artificial Intelligence-Based Brain-Computer Interface, V. Bajaj and G. R. Sinha, Eds. Academic Press, 2022, pp. 145–162.
• [19] S. K. Khare, G. S. Member, V. Bajaj, S. Member, U. R. Acharya, and S. Member, “SPWVD-CNN for Automated Detection of Schizophrenia Patients Using EEG Signals,” vol. 70, 2021.
• [20] A. Vaswani, N. Shazeer, and N. Parmar, “Attention is All You Need,” in 31st Conference on Neural Information Processing Systems (NIPS), 2015.
• [21] A. Dosovitskiy, L. Beyer, A. Kolesnikov, and D. Weissenborn, “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” in ICLR 2021, 2021.