Detection of Attention Deficit Hyperactivity Disorder by Using Eeg Signals and Deep Learning
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
İzmir Ekonomi Üniversitesi
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Dikkat eksikliği hiperaktivite bozukluğu (DEHB) genelikle çocuklarda görülen, nörolojik bir hastalıktır. Bu hastalık yaşam kalitesini düşürdüğü için erken teşhis son derece önemlidir. Elektroensefalografi (EEG) bir nörogörüntüleme tekniği olup DEHB teşhisinde yoğun olarak kullanılmaktadır. Bu çalışmada aynı yaş grubundaki DEHB tanısı konmuş bireyler ve sağlıklı kontrol bireylerden dinlenme durumu EEG sinyaleri İzmir Katip Çelebi Üniversitesi, Nöroloji Kliniğinde kayıt edilmiştir. Görgül Kip Ayrışım (GKA: EMD) yöntemi ile içkin kip fonksiyonları (IMF) elde edilmiştir. Daha sonra IMF'ler ve EEG sinyallerinin kısa süreli Fourier dönüşümü yardımı ile spektrogramları hesaplanmış ve renkli imgeler olarak kayıt edilmiştir. Daha sonra spektrogram imgeleri beynin farklı bölgeleri ve bütünü bir arada kullanılarak iki boyutlu Evrişimsel Sinir Ağları (2D-CNN) eğitilerek sınıflandırılmıştır. Yapılan testlerde Python ortamında tasarlanmış olan CNN yapısı ile yaklaşık % 92, ResNet50 mimarisi ile % 96.526 sınıflandırma doğruluğuna ulaşılmıştır.
Attention deficit hyperactivity disorder (ADHD) is a neurological disorder generally seen in children, and early diagnosis is extremely important. Electroencephalography (EEG) signals are used extensively to diagnose ADHD. In this study, resting state EEG signals from ADHD patients and healthy control subjects in the same age group were recorded at the Izmir Katip Celebi University, Department of Neurology, and analyzed. Intrinsic mode functions (IMF) were extracted by the Empirical Mode Decomposition (EMD) method. Then, short-term Fourier transform spectrograms of IMFs as well as the EEG signals were calculated and saved as colored images. Finally, the spectrogram images were classified by training two-dimensional Convolutional Neural Networks (2D-CNN) using different brain regions or the whole brain. In our simulations, almost 92% classification accuracy was achieved with the CNN structure designed in the Python environment, and 96.526% classification accuracy was achieved with the ResNet50 architecture.
Attention deficit hyperactivity disorder (ADHD) is a neurological disorder generally seen in children, and early diagnosis is extremely important. Electroencephalography (EEG) signals are used extensively to diagnose ADHD. In this study, resting state EEG signals from ADHD patients and healthy control subjects in the same age group were recorded at the Izmir Katip Celebi University, Department of Neurology, and analyzed. Intrinsic mode functions (IMF) were extracted by the Empirical Mode Decomposition (EMD) method. Then, short-term Fourier transform spectrograms of IMFs as well as the EEG signals were calculated and saved as colored images. Finally, the spectrogram images were classified by training two-dimensional Convolutional Neural Networks (2D-CNN) using different brain regions or the whole brain. In our simulations, almost 92% classification accuracy was achieved with the CNN structure designed in the Python environment, and 96.526% classification accuracy was achieved with the ResNet50 architecture.
Description
Keywords
Elektrik ve Elektronik Mühendisliği, Electrical and Electronics Engineering
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
N/A
Scopus Q
N/A
Source
Volume
Issue
Start Page
1
End Page
72
