Medical image reader powered by artificial intelligence

This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2023.

書誌詳細
主要な著者:	Palok, Tanvir Ahmed, Ahmed, Symum, Anim, Golam Kibria, Ratul, Shahed Sharif Bhuiyan, Alam, Shahrear
その他の著者:	Nahim, Nabuat Zaman
フォーマット:	学位論文
言語:	English
出版事項:	Brac University 2024
主題:	Misdiagnosis Deep learning Ensemble learning Histogram equalization Transfer learning Data mining Neural networks (Computer science) Ensemble learning (Machine learning) > Industrial applications
オンライン･アクセス:	http://hdl.handle.net/10361/23517

id	10361-23517
record_format	dspace
spelling	10361-235172024-06-23T21:00:23Z Medical image reader powered by artificial intelligence Palok, Tanvir Ahmed Ahmed, Symum Anim, Golam Kibria Ratul, Shahed Sharif Bhuiyan Alam, Shahrear Nahim, Nabuat Zaman Department of Computer Science and Engineering, Brac University Misdiagnosis Deep learning Ensemble learning Histogram equalization Transfer learning Data mining Neural networks (Computer science) Ensemble learning (Machine learning)--Industrial applications This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2023. Cataloged from PDF version of thesis. Includes bibliographical references (pages 42-44). Misdiagnosis in medical imaging is a critical concern, risking patients’ health due to the pivotal role of radiologists’ accuracy in diagnostics. Current cross-checking methods for radiologists’ decisions are limited, potentially leading to errors and treatment delays. This study introduces a data processing technique and an advanced prediction system for improving disease detection accuracy in medical images. Our main goal is to contribute to healthcare by developing a system capable of achieving human-level or higher accuracy in disease detection across diverse medical image types. To achieve this, we utilize deep learning techniques, specifically Convolutional Neural Networks (CNNs), and leverage Transfer Learning with pre-trained models. Data processing plays a crucial role, given the importance of image availability and quality. We apply image enhancement techniques such as Histogram Equalization, Adaptive Histogram Equalization (AHE), and Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance image quality and augment a limited training dataset. The advanced ensemble approach significantly enhances the overall accuracy and reduces individual model variance. Validation of our approach using confusion matrices reveals that selective class-wise voting achieves the highest accuracy at 95.27% on the testing dataset. Additionally, our customized weighted voting approach achieves an accuracy of 94.07% on the test set. These results emphasize the effectiveness of our ensemble techniques in improving disease detection accuracy. Our ensemble techniques offer substantial accuracy improvements, promising more accurate and reliable medical diagnoses Tanvir Ahmed Palok Symum Ahmed Golam Kibria Anim Shahed Sharif Bhuiyan Ratul Shahrear Alam B.Sc in Computer Science 2024-06-23T10:19:30Z 2024-06-23T10:19:30Z ©2023 2023-09 Thesis ID 19301012 ID 19101456 ID 23341034 ID 23341059 ID 19301016 http://hdl.handle.net/10361/23517 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. 56 pages application/pdf Brac University
institution	Brac University
collection	Institutional Repository
language	English
topic	Misdiagnosis Deep learning Ensemble learning Histogram equalization Transfer learning Data mining Neural networks (Computer science) Ensemble learning (Machine learning)--Industrial applications
spellingShingle	Misdiagnosis Deep learning Ensemble learning Histogram equalization Transfer learning Data mining Neural networks (Computer science) Ensemble learning (Machine learning)--Industrial applications Palok, Tanvir Ahmed Ahmed, Symum Anim, Golam Kibria Ratul, Shahed Sharif Bhuiyan Alam, Shahrear Medical image reader powered by artificial intelligence
description	This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2023.
author2	Nahim, Nabuat Zaman
author_facet	Nahim, Nabuat Zaman Palok, Tanvir Ahmed Ahmed, Symum Anim, Golam Kibria Ratul, Shahed Sharif Bhuiyan Alam, Shahrear
format	Thesis
author	Palok, Tanvir Ahmed Ahmed, Symum Anim, Golam Kibria Ratul, Shahed Sharif Bhuiyan Alam, Shahrear
author_sort	Palok, Tanvir Ahmed
title	Medical image reader powered by artificial intelligence
title_short	Medical image reader powered by artificial intelligence
title_full	Medical image reader powered by artificial intelligence
title_fullStr	Medical image reader powered by artificial intelligence
title_full_unstemmed	Medical image reader powered by artificial intelligence
title_sort	medical image reader powered by artificial intelligence
publisher	Brac University
publishDate	2024
url	http://hdl.handle.net/10361/23517
work_keys_str_mv	AT paloktanvirahmed medicalimagereaderpoweredbyartificialintelligence AT ahmedsymum medicalimagereaderpoweredbyartificialintelligence AT animgolamkibria medicalimagereaderpoweredbyartificialintelligence AT ratulshahedsharifbhuiyan medicalimagereaderpoweredbyartificialintelligence AT alamshahrear medicalimagereaderpoweredbyartificialintelligence
_version_	1814306893514932224

Medical image reader powered by artificial intelligence

類似資料