Diagnosis of schizophrenia based on transformation from EEG sub-bands to the image with deep learning architecture

Basit Öğe Kaydı
dc.authorid0000-0002-0060-1880en_US
dc.authorscopusid57195215516en_US
dc.contributor.authorTürk, Ömer
dc.contributor.authorAldemir, Erdoğan
dc.contributor.authorAcar, Emrullah
dc.contributor.authorErtuğrul, Ömer Faruk
dc.date.accessioned2024-01-02T08:13:42Z
dc.date.available2024-01-02T08:13:42Z
dc.date.issued2023en_US
dc.departmentMAÜ, Fakülteler, Mühendislik Mimarlık Fakültesi, Bilgisayar Mühendisliği Bölümüen_US
dc.description.abstractElectroencephalogram is a low-cost, non-invasive, and high-entropy signal and thus has huge potential for clinical diagnosis of neurological diseases and brain–computer interface applications. Schizophrenia is one of the most severe diseases that show behavioral manifestations that are easily uncovered by specialists. In this context, the electroencephalogram analysis becomes more important for the automatic diagnosis of schizophrenia disease in the clinical process. In this study, a deep learning architecture, namely ResNet, aims to classify schizophrenia is proposed. The proposed system transforms wavelet sub-bands of the electroencephalogram into two-dimensional image space, which is considered the main unique contribution of the study. Thus, the disease indicators and features included in images could be figured out. Moreover, a discussion on the class activation maps was made to give a wide perspective on the features related to the disease. The proposed system was implemented on a large-scale electroencephalogram database containing records from unhealthy and healthy patients in various phases. The ResNet was implemented in three modes to give a thorough perspective in terms of the metrics of the diagnosis accuracy. The proposed system achieves 92.94% diagnosis accuracy rate, and the result shows that the proposed transformation-based solution is owing to the features related to schizophrenia diseaseen_US
dc.identifier.citationTÜRK, Ö., ALDEMİR, E., ACAR, E., & ERTUĞRUL, Ö. F. (2023). Diagnosis of schizophrenia based on transformation from EEG sub-bands to the image with deep learning architecture. SOFT COMPUTING, 0–0.en_US
dc.identifier.doi10.1007/s00500-023-09492-zen_US
dc.identifier.scopus2-s2.0-85180443484en_US
dc.identifier.scopusqualityN/Aen_US
dc.identifier.urihttps://doi.org/10.1007/s00500-023-09492-z
dc.identifier.urihttps://hdl.handle.net/20.500.12514/5357
dc.identifier.wosWOS:001130380600005en_US
dc.identifier.wosqualityN/Aen_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.institutionauthorTürk, Ömer
dc.language.isoenen_US
dc.publisherSpringer Science and Business Media Deutschland GmbHen_US
dc.relation.ispartofSoft Computingen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectAutomatic diagnosisen_US
dc.subjectDeep learningen_US
dc.subjectEEGen_US
dc.subjectSchizophreniaen_US
dc.subjectTransformationen_US
dc.titleDiagnosis of schizophrenia based on transformation from EEG sub-bands to the image with deep learning architectureen_US
dc.typeArticleen_US

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
turk.pdf
Boyut:
2.37 MB
Biçim:
Adobe Portable Document Format
Açıklama:
Article/Makale
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Scopus İndeksli Yayınlar Koleksiyonu
Bilgisayar Mühendisliği Bölümü Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu