This paper concerns the diagnosis of schizophrenia using encephalographic signals and introduces a new framework based on image processing technique. Time-frequency approach or spectrogram image processing technique was used to analyze EEG signals. The spectrogram images were formed from EEG signals, then the Gray Level Co-occurrence Matrix (GLCM) texture feature was extracted from the images. This texture feature produced huge matrix data, thus we used locally linear embedding algorithm (LLA) to reduce the big matrix. In this model, the neuro-based computational model on the limbic system was used to discriminate subjects with schizophrenia patients and control participant that models the emotional process. This architecture is a merging algorithm based on brain emotional learning and fuzzy inference system. The results showed that the proposed model is able to classify the electroencephalographic spectrogram image with 81.5% accuracy.
|Title of host publication||2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2017|
|Publisher||IEEE, Institute of Electrical and Electronics Engineers|
|Publication status||Published - 4 Oct 2017|
|Event||2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2017 - Manchester, United Kingdom|
Duration: 23 Aug 2017 → 25 Aug 2017
|Conference||2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2017|
|Period||23/08/17 → 25/08/17|