Deep Convolutional Neural Network Model for the Differential Diagnosis of Schizophrenia Using EEG Signals

Filiz Demirdöğen; Çağla Danacı; Seda Arslan Tuncer; Mustafa Akkuş; Sevler Yıldız

Makale detayları

Deep Convolutional Neural Network Model for the Differential Diagnosis of Schizophrenia Using EEG Signals

Yazarlar:

Filiz Demirdöğen | Çağla Danacı | Seda Arslan Tuncer | Mustafa Akkuş | Sevler Yıldız

Yayıncı:

Hitit University

Ana Başlıklar:

Anesteziyoloji

Dergi:

Hitit Medical Journal Vol.6 Issue 3, pp.257-265

Belge Tipi:

Makale

Yayın Tarihi:

10/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001329535

pISSN:

2687-4717

Anahtar kelimeler:

Özet:

 Objective: One of the serious mental disorders in which people interpret reality in an abnormal situation is schizophrenia. A combination of extremely disordered thoughts, delusions, and hallucinations occurs due to schizophrenia, and the persons daily functions are seriously impaired due to this disease. For general cognitive activity analysis, electroencephalography signals are widely used as a low-resolution diagnostic tool. This study aimed to diagnose schizophrenia using the transfer learning method by including the EEGs of 73 patients diagnosed with schizophrenia, and 67 patients from the healthy group. Material and Method: In the first step of the study, digital electroencephalography signal data was converted into spectrograms to make them usable. In the classification phase, ResNet18, ResNet50 and EfficientNet models, which are FastAI, and Convolutional Neural Network CNN based deep learning models, were used. Results: Despite the complexity of electroencephalography data, CNN-based models in the study were successful in capturing different aspects of neurophysiological activity. The best performance was obtained from the ResNet-50 model with an accuracy rate of 97. Afterwards, the classification process was finalized with 95 ResNet-18, and 83 EfficientNet models, respectively. Conclusion: It is thought that the classification performance of the result obtained in the application is promising, and may be a guide for future studies.

 Objective: One of the serious mental disorders in which people interpret reality in an abnormal situation is schizophrenia. A combination of extremely disordered thoughts, delusions, and hallucinations occurs due to schizophrenia, and the persons daily functions are seriously impaired due to this disease. For general cognitive activity analysis, electroencephalography signals are widely used as a low-resolution diagnostic tool. This study aimed to diagnose schizophrenia using the transfer learning method by including the EEGs of 73 patients diagnosed with schizophrenia, and 67 patients from the healthy group. Material and Method: In the first step of the study, digital electroencephalography signal data was converted into spectrograms to make them usable. In the classification phase, ResNet18, ResNet50 and EfficientNet models, which are FastAI, and Convolutional Neural Network CNN based deep learning models, were used. Results: Despite the complexity of electroencephalography data, CNN-based models in the study were successful in capturing different aspects of neurophysiological activity. The best performance was obtained from the ResNet-50 model with an accuracy rate of 97. Afterwards, the classification process was finalized with 95 ResNet-18, and 83 EfficientNet models, respectively. Conclusion: It is thought that the classification performance of the result obtained in the application is promising, and may be a guide for future studies.

 Objective: One of the serious mental disorders in which people interpret reality in an abnormal situation is schizophrenia. A combination of extremely disordered thoughts, delusions, and hallucinations occurs due to schizophrenia, and the persons daily functions are seriously impaired due to this disease. For general cognitive activity analysis, electroencephalography signals are widely used as a low-resolution diagnostic tool. This study aimed to diagnose schizophrenia using the transfer learning method by including the EEGs of 73 patients diagnosed with schizophrenia, and 67 patients from the healthy group. Material and Method: In the first step of the study, digital electroencephalography signal data was converted into spectrograms to make them usable. In the classification phase, ResNet18, ResNet50 and EfficientNet models, which are FastAI, and Convolutional Neural Network CNN based deep learning models, were used. Results: Despite the complexity of electroencephalography data, CNN-based models in the study were successful in capturing different aspects of neurophysiological activity. The best performance was obtained from the ResNet-50 model with an accuracy rate of 97. Afterwards, the classification process was finalized with 95 ResNet-18, and 83 EfficientNet models, respectively. Conclusion: It is thought that the classification performance of the result obtained in the application is promising, and may be a guide for future studies.

 Amaç: İnsanların gerçekliği anormal bir durumda yorumladığı ciddi zihinsel bozukluklardan biri de şizofrenidir. Şizofreni nedeniyle aşırı derecede düzensiz düşünce, sanrı ve halüsinasyonların birleşimi ortaya çıkmakta ve bu hastalık nedeniyle kişinin günlük işlevleri ciddi şekilde bozulmaktadır. Genel bilişsel aktivite analizi için elektroensefalografi sinyalleri, düşük çözünürlüklü bir teşhis aracı olarak yaygın olarak kullanılmaktadır. Bu çalışmaya şizofreni tanısı almış 73 hasta ile sağlıklı grubuna ait 67 hastanın EEGsi dahil edilerek transfer öğrenme metodu ile şizofreni teşhisi gerçekleştirmek amaçlanmıştır. Gereç ve Yöntem: Çalışmanın ilk adımında sayısal elektroensefalografi sinyal verilerini kullanılabilir hale getirmek amacıyla spektogramlara dönüştürme işlemi gerçekleştirilmiştir. Sınıflandırma aşamasında FastAI ile Convolutional Neural Network CNN tabanlı derin öğrenme modelleri olan ResNet18, ResNet50 ve EfficientNet modelleri kullanılmıştır. Bulgular: Elektroensefalografi verilerinin karmaşıklığına rağmen çalışmada CNN tabanlı modeller, nörofizyolojik aktivitenin farklı yönlerini yakalamada başarılı olmuştur. En iyi performans 97 doğruluk oranı ile ResNet-50 modelinden elde edilmiştir. Sonrasında sırasıyla 95 ResNet-18 ve 83 EfficientNet modelleri ile sınıflandırma işlemi sonuçlandırılmıştır. Sonuç: Uygulamada ulaşılan sonucun sınıflandırma performansının umut verici olduğu ve bundan sonraki yapılacak çalışmalar için yol gösterici nitelikte olabileceği düşünülmektedir.

Makale detayları

Deep Convolutional Neural Network Model for the Differential Diagnosis of Schizophrenia Using EEG Signals

Yazarlar:

Filiz Demirdöğen | Çağla Danacı | Seda Arslan Tuncer | Mustafa Akkuş | Sevler Yıldız

Yayıncı:

Hitit University

Ana Başlıklar:

Anesteziyoloji

Dergi:

Hitit Medical Journal Vol.6 Issue 3, pp.257-265

Belge Tipi:

Makale

Yayın Tarihi:

10/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001329535

pISSN:

2687-4717

Artificial Intelligence

EEG

FastAI

Schizophrenia

Transfer Learning.FastAI

EEG

Transfer Öğrenme

Şizofreni

Yapay Zekâ

Sayı

Makale detayları

Yazarlar: Filiz Demirdöğen | Çağla Danacı | Seda Arslan Tuncer | Mustafa Akkuş | Sevler Yıldız

Yayıncı: Hitit University

Ana Başlıklar: Anesteziyoloji

Dergi: Hitit Medical Journal Vol.6 Issue 3, pp.257-265

Belge Tipi: Makale

Yayın Tarihi: 10/2024

Belge Dili: İngilizce

Document ID (Org No.): 2001329535

pISSN: 2687-4717

Sayı

Yazarlar:

Filiz Demirdöğen | Çağla Danacı | Seda Arslan Tuncer | Mustafa Akkuş | Sevler Yıldız

Yayıncı:

Hitit University

Ana Başlıklar:

Anesteziyoloji

Dergi:

Hitit Medical Journal Vol.6 Issue 3, pp.257-265

Belge Tipi:

Makale

Yayın Tarihi:

10/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001329535

pISSN:

2687-4717