Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network

This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Electrical and Electronic Engineering, 2020.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Κύριοι συγγραφείς:	Ismail, Sayem Mohammad, Hossain, Md. Sajjad, Sobhan, Irina
Άλλοι συγγραφείς:	Bhuian, Mohammed Belal Hossain
Μορφή:	Thesis
Γλώσσα:	en_US
Έκδοση:	Brac University 2021
Θέματα:	Glaucoma Artiﬁcial Intelligence Fundus images CNN Transfer leaning Prediction Deep learning VGG19 ResNet50 DenseNet121 InceptionV3 Finetuning
Διαθέσιμο Online:	http://hdl.handle.net/10361/14365

id	10361-14365
record_format	dspace
spelling	10361-143652021-03-21T21:01:24Z Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network Ismail, Sayem Mohammad Hossain, Md. Sajjad Sobhan, Irina Bhuian, Mohammed Belal Hossain Alam, Md. Ashraful Department of Electrical and Electronic Engineering, Brac University Glaucoma Artiﬁcial Intelligence Fundus images CNN Transfer leaning Prediction Deep learning VGG19 ResNet50 DenseNet121 InceptionV3 Finetuning This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Electrical and Electronic Engineering, 2020. Cataloged from PDF version of thesis. Includes bibliographical references (pages 52-55). Transfer learning techniques in deep learning is nowadays a raising and promising ﬁeld of research and a tool for Artiﬁcial Intelligence with a lot of prospects. Our goal is to predict Glaucoma from fundus images to help the diagnosis procedure of Glaucoma, a public health hazard, at an early stage. In this research, we propose a transfer leaning methodology, creating four models with four pre-trained CNNs implemented separately in each of the models, trained and tested for detecting Glaucoma from fundus images. We have used VGG19, ResNet50, DenseNet121 and InceptionV3 for our transfer learning models, with the ﬁne-tuning approach to ensure better learning performance on our dataset of labelled fundus images. Fine-tuning is done keeping all the layers of pre-trained CNN trainable on the fundus image dataset, and applying the classic method of adding a customized classiﬁer. All the four transfer leaning models are Deep Neural Networks carrying deep hidden layers as the pre-trained CNN implemented. Deep learning application on Biomedical ﬁeld is itself a challenge to work with due to shortage of labeled data. Thus transfer learning is found very eﬀective in working with a small image data set to predict Glaucoma. Our proposed models built with VGG19, ResNet50, DenseNet121 and InceptionV3 deliver test accuracy of 94.75%, 96.5%, 92.5%, 91.75%. In order to achieve such accuracy in biomedical application, transfer of knowledge of features learned of pre-trained CNNs gave a competitive edge on initialization of parameters. We present comparison amongst the models proposed and the ResNet50 built model gives the best performance. Sayem Mohammad Ismail Md. Sajjad Hossain Irina Sobhan B. Electrical and Electronic Engineering 2021-03-21T07:03:13Z 2021-03-21T07:03:13Z 2020 2020-06 Thesis ID: 16321050 ID: 16321062 ID: 16121133 http://hdl.handle.net/10361/14365 en_US 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. 55 pages application/pdf Brac University
institution	Brac University
collection	Institutional Repository
language	en_US
topic	Glaucoma Artiﬁcial Intelligence Fundus images CNN Transfer leaning Prediction Deep learning VGG19 ResNet50 DenseNet121 InceptionV3 Finetuning
spellingShingle	Glaucoma Artiﬁcial Intelligence Fundus images CNN Transfer leaning Prediction Deep learning VGG19 ResNet50 DenseNet121 InceptionV3 Finetuning Ismail, Sayem Mohammad Hossain, Md. Sajjad Sobhan, Irina Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
description	This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Electrical and Electronic Engineering, 2020.
author2	Bhuian, Mohammed Belal Hossain
author_facet	Bhuian, Mohammed Belal Hossain Ismail, Sayem Mohammad Hossain, Md. Sajjad Sobhan, Irina
format	Thesis
author	Ismail, Sayem Mohammad Hossain, Md. Sajjad Sobhan, Irina
author_sort	Ismail, Sayem Mohammad
title	Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
title_short	Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
title_full	Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
title_fullStr	Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
title_full_unstemmed	Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
title_sort	prediction of glaucoma from fundus images leveraging transfer learning in deep neural network
publisher	Brac University
publishDate	2021
url	http://hdl.handle.net/10361/14365
work_keys_str_mv	AT ismailsayemmohammad predictionofglaucomafromfundusimagesleveragingtransferlearningindeepneuralnetwork AT hossainmdsajjad predictionofglaucomafromfundusimagesleveragingtransferlearningindeepneuralnetwork AT sobhanirina predictionofglaucomafromfundusimagesleveragingtransferlearningindeepneuralnetwork
_version_	1814307345890541568

Prediction of glaucoma from fundus images leveraging transfer learning in deep neural network

Παρόμοια τεκμήρια