A GAN-based federated learning architecture for data augmentation of medical images

This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022.

Sonraí bibleagrafaíochta
Príomhchruthaitheoirí:	Al Rakin, Abdullah, Iqbal Majumder, MD. Akib, Kabir, Mohammad Farhan, Arafin, Rudmila
Rannpháirtithe:	Reza, Tanzim
Formáid:	Tráchtas
Teanga:	English
Foilsithe / Cruthaithe:	Brac University 2023
Ábhair:	GAN Generator Discriminator Federated Learning OCT Deep Convolutional Generative Adversarial Network (DCGAN) Wasserstein GAN (WGAN) Distributed GAN Mode Collapse Non-iid Data. Natural language processing (Computer science) Machine learning
Rochtain ar líne:	http://hdl.handle.net/10361/18089

id	10361-18089
record_format	dspace
spelling	10361-180892023-04-05T21:01:52Z A GAN-based federated learning architecture for data augmentation of medical images Al Rakin, Abdullah Iqbal Majumder, MD. Akib Kabir, Mohammad Farhan Arafin, Rudmila Reza, Tanzim Hossain, Dr. Muhammad Iqbal Department of Computer Science and Engineering, Brac University GAN Generator Discriminator Federated Learning OCT Deep Convolutional Generative Adversarial Network (DCGAN) Wasserstein GAN (WGAN) Distributed GAN Mode Collapse Non-iid Data. Natural language processing (Computer science) Machine learning This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022. Cataloged from PDF version of thesis. Includes bibliographical references (pages 30-32). n the medical industry, the availability of precise data limits the scope of deep learn ing applications. Institutional norms restrict hospitals and research facilities owing to privacy concerns. Therefore, data collection from such sources is unfeasible. Fed erated Learning (FL) is promising in this scenario, but it does not guarantee data privacy. In this paper, we will use Deep Convolutional Generative Adversarial Net work (DCGAN) and Wasserstein Generative Adversarial Network (WGAN) on an OCT dataset to demonstrate that the Federated GAN (FedGAN) architecture fails in these networks due to its innate structure. Additionally, introduce a Distributed Generative Adversarial Network (Distributed GAN) that collects and distributes the weights of each temporary GANs on the client side to the main server to tackle the mode collapse risk of non-iid data. This conserves the optimal distribution of data to all private discriminators while protecting sensitive individual data. Abdullah Al Rakin MD. Akib Iqbal Majumder Mohammad Farhan Kabir Rudmila Arafin B. Computer Science 2023-04-05T09:13:05Z 2023-04-05T09:13:05Z 2022 2022-09 Thesis ID: 21341032 ID: 18201142 ID: 19101530 ID: 18301105 http://hdl.handle.net/10361/18089 en Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 32 pages application/pdf Brac University
institution	Brac University
collection	Institutional Repository
language	English
topic	GAN Generator Discriminator Federated Learning OCT Deep Convolutional Generative Adversarial Network (DCGAN) Wasserstein GAN (WGAN) Distributed GAN Mode Collapse Non-iid Data. Natural language processing (Computer science) Machine learning
spellingShingle	GAN Generator Discriminator Federated Learning OCT Deep Convolutional Generative Adversarial Network (DCGAN) Wasserstein GAN (WGAN) Distributed GAN Mode Collapse Non-iid Data. Natural language processing (Computer science) Machine learning Al Rakin, Abdullah Iqbal Majumder, MD. Akib Kabir, Mohammad Farhan Arafin, Rudmila A GAN-based federated learning architecture for data augmentation of medical images
description	This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2022.
author2	Reza, Tanzim
author_facet	Reza, Tanzim Al Rakin, Abdullah Iqbal Majumder, MD. Akib Kabir, Mohammad Farhan Arafin, Rudmila
format	Thesis
author	Al Rakin, Abdullah Iqbal Majumder, MD. Akib Kabir, Mohammad Farhan Arafin, Rudmila
author_sort	Al Rakin, Abdullah
title	A GAN-based federated learning architecture for data augmentation of medical images
title_short	A GAN-based federated learning architecture for data augmentation of medical images
title_full	A GAN-based federated learning architecture for data augmentation of medical images
title_fullStr	A GAN-based federated learning architecture for data augmentation of medical images
title_full_unstemmed	A GAN-based federated learning architecture for data augmentation of medical images
title_sort	gan-based federated learning architecture for data augmentation of medical images
publisher	Brac University
publishDate	2023
url	http://hdl.handle.net/10361/18089
work_keys_str_mv	AT alrakinabdullah aganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT iqbalmajumdermdakib aganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT kabirmohammadfarhan aganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT arafinrudmila aganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT alrakinabdullah ganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT iqbalmajumdermdakib ganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT kabirmohammadfarhan ganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages AT arafinrudmila ganbasedfederatedlearningarchitecturefordataaugmentationofmedicalimages
_version_	1814309506916548608

A GAN-based federated learning architecture for data augmentation of medical images

Míreanna comhchosúla