|
|
EMU I-REP >
02 Faculty of Engineering >
Department of Computer Engineering >
Theses (Master's and Ph.D) – Computer Engineering >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/11129/6551
|
| Title: | Deep Learning Based Processing of EEG Signals for Detection and Recognition of Parkinson Disease |
| Authors: | Acan, Adnan Reyhanlı, Samed Eastern Mediterranean University, Faculty of Engineering, Dept. of Computer Engineering |
| Keywords: | Thesis Tez Computer Engineering Department Artificial intelligence--Medical applications Medical informatics Biomedical engineering Artificial Intelligence--Medical Applications Artificial Intelligence--Computational intelligence Parkinson disease deep learning eeg gasf cnn |
| Issue Date: | Aug-2022 |
| Publisher: | Eastern Mediterranean University (EMU) - Doğu Akdeniz Üniversitesi (DAÜ) |
| Citation: | Reyhanlı, Samed. (2022).Deep Learning Based Processing of EEG Signals for Detection and Recognition of Parkinson Disease. Thesis (M.S.), Eastern Mediterranean University, Institute of Graduate Studies and Research, Dept. of Computer Engineering, Famagusta: North Cyprus. |
| Abstract: | The aim of this study is to provide early detection of Parkinson's disease by processing
EEG signals through two dimensional colored image transforms.
Parkinson's disease is a neurological disease that usually occurs in old ages and occurs
with a decrease in dopamine levels in the brain. There is no known treatment for
Parkinson's disease. Early detection and early treatment in Parkinson's disease is very
important to slow the progression of the disease.
EEG data were obtained from the UC San Diego Resting State EEG Database from
Patients with Parkinson's disease. EEG signals were converted to GASF images by
going through various preprocessing steps.
AlexNet deep learning model was used to train and test the obtained 2D colored image
data. AlexNet is a Convolutional Neural Network model consisting of 8 layers. In the
literature review, 16 channels used in various studies were selected. Amoung these
Fp1, F7 and F3 channels are the ones with highest reported succes results. The same
channels are also considered with in the scope of address in this thesis work.
GASF images of selected Fp1, F7 and F3 channels were used to train the AlexNet
CNN model over 100 epochs. The developed model achieved promising performance
with 97.72% accuracy, 97.76% sensitivity and 97.68% specificity. In addition, the
AlexNet CNN model was trained and tested over 100 epochs with 4-fold Cross
Validation.
As a result of this study, the developed model achieved the highest results with 97.73%
accuracy, 97.94% sensitivity and 97.53% specificity. ÖZ:
Bu çalışmanın amacı, iki boyutlu renkli görüntü dönüşümlerini işleyerek Parkinson
hastalığının erken tespitini sağlamaktır.
Parkinson hastalığı, genellikle ileri yaşlarda ortaya çıkan ve beyindeki dopamin
seviyesinin düşmesiyle ortaya çıkan nörolojik bir hastalıktır. Parkinson hastalığının
bilinen bir tedavisi yoktur. Parkinson hastalığında erken teşhis ve erken tedavi,
hastalığın ilerlemesini yavaşlatmak için çok önemlidir.
EEG verileri, Parkinson hastalığı olan Hastalardan alınan UC San Diego Dinlenme
Durumu EEG Veritabanından elde edilmiştir. EEG sinyalleri çeşitli ön işleme
adımlarından geçirilerek GASF görüntülerine dönüştürülmüştür.
Elde edilen verileri eğitmek ve test etmek için AlexNet derin öğrenme modeli
kullanıldı. AlexNet, 8 katmandan oluşan bir Evrişimsel Sinir Ağı modelidir. Literatür
taramasında çeşitli çalışmalarda kullanılan 16 kanal seçilmiş ve bu kanallar arasından
en yüksek sonucu veren Fp1, F7 ve F3 kanalları nihai sonuç için kullanılmıştır. Bu
seçim yapılırken her kanal ayrı ayrı eğitilmiş ve test edilmiştir. Elde edilen sonuçlar
analiz edilirken doğruluğu en yüksek olan kanallar seçilmiştir.
Seçilen Fp1, F7 ve F3 kanallarının GASF görüntüleri AlexNet CNN modelinde 100
dönem boyunca eğitilmiştir. Geliştirilen model %97,72 doğruluk, %97,76 duyarlılık
ve %97,68 özgüllük ile umut verici bir performans elde etmiştir. Ayrıca, AlexNet CNN
modeli, k-Fold Cross Validation kullanılarak 4-fold ile 100 epoch boyunca eğitilmiş
ve test edilmiştir. 4-Fold sonucunda geliştirilen model %97.73 doğruluk, %97.94
duyarlılık ve %97.53 özgüllük ile en yüksek sonuçlar elde edilmiştir. |
| Description: | Master of Science in Computer Engineering. Institute of Graduate Studies and Research. Thesis (M.S.) - Eastern Mediterranean University, Faculty of Engineering, Dept. of Computer Engineering, 2022. Supervisor: Assoc. Prof. Dr. Adnan Acan. |
| URI: | http://hdl.handle.net/11129/6551 |
| Appears in Collections: | Theses (Master's and Ph.D) – Computer Engineering
|
This item is protected by original copyright
|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
|