Non-Invasive Bio-Signal Data Classification Of Psychiatric Mood Disorders Using Modified CNN and VGG16

Ali Berkan Ural

Article Details

Non-Invasive Bio-Signal Data Classification Of Psychiatric Mood Disorders Using Modified CNN and VGG16

Authors:

Ali Berkan Ural

Publisher:

Kirikkale University

Main Topics:

Engineering

Journal:

International Journal of Engineering Research and Development Vol.15 Issue 1, pp.323-332

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281381

pISSN:

1308-5506

Keywords:

Abstract:

 In this study, the aim is to develop an ensemble machine learning ML based deep learning DL model classifiers to detect and compare one type of major psychiatric disorders of mood disorders Depressive and Bipolar disorders using Electroencephalography EEG. The diverse and multiple non-invasive biosignals were collected retrospectively according to the granted ethical permission. The experimental part is consisted from three main parts. First part is the data collectiondevelopment, the second part is data transformation and augmentation via Spectrogram image conversion process and online Keras data augmentation part, respectively. The third and final part is to fed these image dataset into modified Convolutional Neural Network CNN and VGG16 models for training and testing parts to detect, compare and discriminate mood disorders types in detail with a specific healthy group. As the performance evaluation background of the mood disorder classification models, confusion matrices and receiver operating characteristics ROC curves were used and finally, the accuracy achieved by CNN model was 88 and VGG16 model was 90, which is an improvement of 10 compared to the previous studies in literature. Therefore, our system can help clinicians and researchers to manage, diagnose and prognosis of the mental health of people.

 In this study, the aim is to develop an ensemble machine learning ML based deep learning DL model classifiers to detect and compare one type of major psychiatric disorders of mood disorders Depressive and Bipolar disorders using Electroencephalography EEG. The diverse and multiple non-invasive biosignals were collected retrospectively according to the granted ethical permission. The experimental part is consisted from three main parts. First part is the data collectiondevelopment, the second part is data transformation and augmentation via Spectrogram image conversion process and online Keras data augmentation part, respectively. The third and final part is to fed these image dataset into modified Convolutional Neural Network CNN and VGG16 models for training and testing parts to detect, compare and discriminate mood disorders types in detail with a specific healthy group. As the performance evaluation background of the mood disorder classification models, confusion matrices and receiver operating characteristics ROC curves were used and finally, the accuracy achieved by CNN model was 88 and VGG16 model was 90, which is an improvement of 10 compared to the previous studies in literature. Therefore, our system can help clinicians and researchers to manage, diagnose and prognosis of the mental health of people.

 In this study, the aim is to develop an ensemble machine learning ML based deep learning DL model classifiers to detect and compare one type of major psychiatric disorders of mood disorders Depressive and Bipolar disorders using Electroencephalography EEG. The diverse and multiple non-invasive biosignals were collected retrospectively according to the granted ethical permission. The experimental part is consisted from three main parts. First part is the data collectiondevelopment, the second part is data transformation and augmentation via Spectrogram image conversion process and online Keras data augmentation part, respectively. The third and final part is to fed these image dataset into modified Convolutional Neural Network CNN and VGG16 models for training and testing parts to detect, compare and discriminate mood disorders types in detail with a specific healthy group. As the performance evaluation background of the mood disorder classification models, confusion matrices and receiver operating characteristics ROC curves were used and finally, the accuracy achieved by CNN model was 88 and VGG16 model was 90, which is an improvement of 10 compared to the previous studies in literature. Therefore, our system can help clinicians and researchers to manage, diagnose and prognosis of the mental health of people.

 Bu çalışmada amaç, Elektroensefalografi EEG kullanarak duygudurum bozukluklarının Depresif ve Bipolar bozukluklar bir tür majör psikiyatrik bozukluğunu saptamak ve karşılaştırmak için topluluk makine öğrenimi ML tabanlı derin öğrenme DL model sınıflandırıcıları geliştirmektir. Çeşitli ve çoklu non-invaziv biyosinyaller, verilen etik izne göre geriye dönük olarak toplandı. Deneysel kısım üç ana bölümden oluşmaktadır. Birinci bölüm veri toplama ve geliştirme, ikinci bölüm ise sırasıyla Spektrogram görüntü dönüştürme işlemi ve çevrimiçi Keras veri artırma bölümü aracılığıyla veri dönüştürme ve artırmadır. Üçüncü ve son bölüm, duygudurum bozuklukları türlerini belirli bir sağlıklı grupla ayrıntılı olarak saptamak, karşılaştırmak ve ayırt etmek için eğitim ve test bölümleri için bu görüntü veri setini değiştirilmiş Konvolüsyonel Sinir Ağı KSA ve VGG16 modellerine beslemektir. Duygudurum bozukluğu sınıflama modellerinin performans değerlendirme arka planı olarak karışıklık matrisleri ve alıcı işletim karakteristikleri ROC eğrileri kullanılmış ve son olarak KSA modelinin doğruluk oranı 88 ve VGG16 modelinin sağladığı doğruluk 90, yani 10luk bir iyileşme olmuştur. literatürdeki önceki çalışmalarla karşılaştırılmıştır. Bu nedenle sistemimiz klinisyenlere ve araştırmacılara insanların ruh sağlığını yönetme, teşhis etme ve tahmin etme konusunda yardımcı olabilir.

Article Details

Non-Invasive Bio-Signal Data Classification Of Psychiatric Mood Disorders Using Modified CNN and VGG16

Authors:

Ali Berkan Ural

Publisher:

Kirikkale University

Main Topics:

Engineering

Journal:

International Journal of Engineering Research and Development Vol.15 Issue 1, pp.323-332

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281381

pISSN:

1308-5506

Psychiatric disorders

Mood disorder

Depressive disorder

Bipolar Disorder

Deep Learning

Pretrained model classification

Psychiatric disorders

Mood disorder

Depressive disorder

Bipolar Disorder

Deep Learning

Pretrained model classification

Issues

Article Details

Authors: Ali Berkan Ural

Publisher: Kirikkale University

Main Topics: Engineering

Journal: International Journal of Engineering Research and Development Vol.15 Issue 1, pp.323-332

Document Type: Article

Publication Date: 1/2023

Document Language: English

Document ID (Org No.): 2001281381

pISSN: 1308-5506

Issues

Authors:

Ali Berkan Ural

Publisher:

Kirikkale University

Main Topics:

Engineering

Journal:

International Journal of Engineering Research and Development Vol.15 Issue 1, pp.323-332

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281381

pISSN:

1308-5506