Show/Hide Menu
Hide/Show Apps
Logout
Türkçe
Türkçe
Search
Search
Login
Login
OpenMETU
OpenMETU
About
About
Open Science Policy
Open Science Policy
Open Access Guideline
Open Access Guideline
Postgraduate Thesis Guideline
Postgraduate Thesis Guideline
Communities & Collections
Communities & Collections
Help
Help
Frequently Asked Questions
Frequently Asked Questions
Guides
Guides
Thesis submission
Thesis submission
MS without thesis term project submission
MS without thesis term project submission
Publication submission with DOI
Publication submission with DOI
Publication submission
Publication submission
Supporting Information
Supporting Information
General Information
General Information
Copyright, Embargo and License
Copyright, Embargo and License
Contact us
Contact us
A comparison of deep neural network architectures for COVID-19 detection using CT chest images
Download
mehmettunahansarioglu_tez.pdf
Date
2022-9
Author
Sarıoğlu, Mehmet Tunahan
Metadata
Show full item record
This work is licensed under a
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
.
Item Usage Stats
250
views
291
downloads
Cite This
Coronavirus Disease 2019 (COVID-19) has rapidly spread around the world since December 2019. Due to the low confidence in the real-time reverse transcription-polymerase chain reaction (RT-PCR) test, which is considered the gold standard, CT images are frequently consulted for diagnosis. CT findings of COVID-19 patients has been analysed and documented in the literature, which are groundglass opacities (GGOs), air bronchograms, vascular enlargement and halo sign. But, these CT findings encountered in COVID-19 disease are not very specific and vary during the course of the disease. In addition, it has common CT appearance and symptoms with many other infectious and non-infectious diseases. This thesis compares various deep transfer learning structures used to distinguish COVID- 19 from other diseases using image processing techniques. In this study, the performances of the classification methods using five different convolutional neural network structures were compared. Unlike other similar studies, two public CT dataset was used together to train, test and validate the methods. Overfitting issues with the train data had been experienced and best scores are obtained using augmented data. Accuracy values of 81.65%, 86.08%, 90.82%, 79.75% and 81.01% were obtained in Xception, VGG 16, ResNet 50, Inception v3 and Inception ResNet v2 convolutional neural networks, respectively. The importance of hyper parameters such as epoch number, loss and activation functions used during training is also mentioned in this study.
Subject Keywords
CT
,
COVID-19
,
Neural Networks
,
Transfer Learning
,
Classification
URI
https://hdl.handle.net/11511/99532
Collections
Graduate School of Informatics, Thesis
Suggestions
OpenMETU
Core
A PROBABILISTIC ASSESSMENT OF SARS-COV-2 HOST INTERACTIONS IN THE CONTEXT OF META-COMMUNITY AND URBAN ECOLOGY
Nehri, Leman Nur; Eroğlu, Seçkin; Department of Biology (2021-6-29)
Coronavirus disease 2019 (COVID-19) is an ongoing pandemic that was detected in Wuhan, China, in December 2019 and spread all around the world. COVID-19 is caused by Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which spreads through close contact. Many studies have been conducted on the transmission, virulence, and immune response of SARS-CoV-2. The intracellular mechanism of action of the virus and various host interaction pathways are also known. In addition, there are many studies on the...
Network intrusion detection system with incremental active learning
Bedir Tüzün, Münteha Nur; Angın, Pelin; Department of Computer Engineering (2022-9-14)
While Internet usage has increased every year, it has gained momentum in recent years with the global pandemic. Increasing Internet usage has brought increasing cyber threats. Intrusion detection systems have become more important than ever. The performance of these systems is directly proportional to their adaptiveness to the rapid changes in attack types. However, desired performance cannot always be achieved due to the lack of labeled data on newly developed attacks and the difficulty of incremental lear...
A New Method Based on Adaptive Discrete Wavelet Entropy Energy and Neural Network Classifier (ADWEENN) for Recognition of Urine Cells from Microscopic Images Independent of Rotation and Scaling
Avcı, Derya; Leblebicioğlu, Mehmet Kemal; POYRAZ, MUSTAFA; DOĞANTEKİN, ESİN (2014-02-01)
So far, analysis and classification of urine cells number has become an important topic for medical diagnosis of some diseases. Therefore, in this study, we suggest a new technique based on Adaptive Discrete Wavelet Entropy Energy and Neural Network Classifier (ADWEENN) for Recognition of Urine Cells from Microscopic Images Independent of Rotation and Scaling. Some digital image processing methods such as noise reduction, contrast enhancement, segmentation, and morphological process are used for feature ext...
Modeling and Characterization of Nanomedicine Transport within Tumor Microenvironment across Scales
Akalın, Ali Aykut; Özçelikkale, Altuğ (2022-09-19)
Despite significant advances in recent decades, diagnosis and treatment of cancer remain to be a major challenge. Advances in nanotechnology have enabled numerous nanoparticle (NP) formulations for efficient delivery of drugs and diagnostic agents to the target tumor site. The size, shape, charge, and surface characteristics of so-called nanomedicine can be tuned to affect the interactions of the NP with the physiological environment [1]. However, design efforts to develop nanomedicine for efficient deliver...
An analysis of the prevalence of malaria in Turkey over the last 85 years
Piyal, Birgul; Akdur, Recep; Ocaktan, Esin; Yozgatlıgil, Ceylan (Informa UK Limited, 2013-01-01)
Background: Affecting 106 countries, malaria is a major global burden. Though intensive antimalaria efforts in Turkey have been successful in bringing down the number of cases, historically malaria was a serious public health concern.
Citation Formats
IEEE
ACM
APA
CHICAGO
MLA
BibTeX
M. T. Sarıoğlu, “A comparison of deep neural network architectures for COVID-19 detection using CT chest images,” M.S. - Master of Science, Middle East Technical University, 2022.