No Cover Image

Journal article 283 views 38 downloads

Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging

Arslan Akbar Orcid Logo, Suya Han Orcid Logo, Naveed Urr Rehman Orcid Logo, Kanwal Ahmed, Hassan Eshkiki Orcid Logo, Fabio Caraffini Orcid Logo

Applied Intelligence, Volume: 55, Issue: 13

Swansea University Authors: Hassan Eshkiki Orcid Logo, Fabio Caraffini Orcid Logo

  • 70043.VoR.pdf

    PDF | Version of Record

    © The Author(s) 2025. This article is licensed under a Creative Commons Attribution 4.0 International License.

    Download (4.99MB)

Check full text

DOI (Published version): 10.1007/s10489-025-06803-9

Abstract

Deep learning models have been instrumental in extracting critical indicators for breast cancer diagnosis - the prevalent malignancy among women worldwide - from baseline magnetic resonance imaging. However, many existing models do not fully leverage the rich spatial information available in the 3D...

Full description

Published in: Applied Intelligence
ISSN: 0924-669X 1573-7497
Published: Springer Science and Business Media LLC 2025
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa70043
first_indexed 2025-07-28T10:16:56Z
last_indexed 2025-08-19T22:12:59Z
id cronfa70043
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-08-18T13:46:46.7654816</datestamp><bib-version>v2</bib-version><id>70043</id><entry>2025-07-28</entry><title>Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging</title><swanseaauthors><author><sid>c9972b26a83de11ffe211070f26fe16b</sid><ORCID>0000-0001-7795-453X</ORCID><firstname>Hassan</firstname><surname>Eshkiki</surname><name>Hassan Eshkiki</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>d0b8d4e63d512d4d67a02a23dd20dfdb</sid><ORCID>0000-0001-9199-7368</ORCID><firstname>Fabio</firstname><surname>Caraffini</surname><name>Fabio Caraffini</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-07-28</date><deptcode>MACS</deptcode><abstract>Deep learning models have been instrumental in extracting critical indicators for breast cancer diagnosis - the prevalent malignancy among women worldwide - from baseline magnetic resonance imaging. However, many existing models do not fully leverage the rich spatial information available in the 3D structure of medical imaging data, potentially overlooking important contextual details. This develops an explainable deep learning framework for classifying breast cancer that leverages the complete 3D and provides classification results alongside visual explanations of the decision-making process. The preprocessing pipeline is fed with 3D sequences containing &#x2018;tumour&#x2019; and &#x2018;non-tumour&#x2019; regions. It includes a 3D Adaptive Unsharp Mask (AUM) filter to reduce noise and augment image class, followed by normalisation and data augmentation. Classification is then achieved by training an augmented ResNet150 model. Three explainable artificial intelligence (XAI) techniques, including Shapley Additive Explanations, 3D Gradient-Weighted Class Activation Mapping, and Contextual Importance and Utility, are employed to provide improved interpretability. The model demonstrates state-of-the-art performance over the QIN-BREAST dataset, achieving testing accuracies of 98.861% for &#x2018;tumours&#x2019; and 99.447% for &#x2018;non-tumours&#x2019;, as well as over the Duke Breast Cancer Dataset, where it achieves 99.104% for &#x2018;tumours&#x2019; and 99.753% for &#x2018;non-tumours&#x2019;, while offering enhanced interpretability through XAI methods.</abstract><type>Journal Article</type><journal>Applied Intelligence</journal><volume>55</volume><journalNumber>13</journalNumber><paginationStart/><paginationEnd/><publisher>Springer Science and Business Media LLC</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0924-669X</issnPrint><issnElectronic>1573-7497</issnElectronic><keywords>Breast cancer; Deep learning; DCE-MRI; Explainable AI; RESNET150</keywords><publishedDay>1</publishedDay><publishedMonth>8</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-08-01</publishedDate><doi>10.1007/s10489-025-06803-9</doi><url/><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders>Swansea University</funders><projectreference/><lastEdited>2025-08-18T13:46:46.7654816</lastEdited><Created>2025-07-28T11:13:44.4778593</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Arslan</firstname><surname>Akbar</surname><orcid>0009-0007-7102-8229</orcid><order>1</order></author><author><firstname>Suya</firstname><surname>Han</surname><orcid>0009-0003-9075-8818</orcid><order>2</order></author><author><firstname>Naveed Urr</firstname><surname>Rehman</surname><orcid>0009-0008-5704-9071</orcid><order>3</order></author><author><firstname>Kanwal</firstname><surname>Ahmed</surname><order>4</order></author><author><firstname>Hassan</firstname><surname>Eshkiki</surname><orcid>0000-0001-7795-453X</orcid><order>5</order></author><author><firstname>Fabio</firstname><surname>Caraffini</surname><orcid>0000-0001-9199-7368</orcid><order>6</order></author></authors><documents><document><filename>70043__34964__914a4dd4307146fb82f38e57c45590ee.pdf</filename><originalFilename>70043.VoR.pdf</originalFilename><uploaded>2025-08-18T13:43:09.9931560</uploaded><type>Output</type><contentLength>5234819</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; The Author(s) 2025. This article is licensed under a Creative Commons Attribution 4.0 International License.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-08-18T13:46:46.7654816 v2 70043 2025-07-28 Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging c9972b26a83de11ffe211070f26fe16b 0000-0001-7795-453X Hassan Eshkiki Hassan Eshkiki true false d0b8d4e63d512d4d67a02a23dd20dfdb 0000-0001-9199-7368 Fabio Caraffini Fabio Caraffini true false 2025-07-28 MACS Deep learning models have been instrumental in extracting critical indicators for breast cancer diagnosis - the prevalent malignancy among women worldwide - from baseline magnetic resonance imaging. However, many existing models do not fully leverage the rich spatial information available in the 3D structure of medical imaging data, potentially overlooking important contextual details. This develops an explainable deep learning framework for classifying breast cancer that leverages the complete 3D and provides classification results alongside visual explanations of the decision-making process. The preprocessing pipeline is fed with 3D sequences containing ‘tumour’ and ‘non-tumour’ regions. It includes a 3D Adaptive Unsharp Mask (AUM) filter to reduce noise and augment image class, followed by normalisation and data augmentation. Classification is then achieved by training an augmented ResNet150 model. Three explainable artificial intelligence (XAI) techniques, including Shapley Additive Explanations, 3D Gradient-Weighted Class Activation Mapping, and Contextual Importance and Utility, are employed to provide improved interpretability. The model demonstrates state-of-the-art performance over the QIN-BREAST dataset, achieving testing accuracies of 98.861% for ‘tumours’ and 99.447% for ‘non-tumours’, as well as over the Duke Breast Cancer Dataset, where it achieves 99.104% for ‘tumours’ and 99.753% for ‘non-tumours’, while offering enhanced interpretability through XAI methods. Journal Article Applied Intelligence 55 13 Springer Science and Business Media LLC 0924-669X 1573-7497 Breast cancer; Deep learning; DCE-MRI; Explainable AI; RESNET150 1 8 2025 2025-08-01 10.1007/s10489-025-06803-9 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University SU Library paid the OA fee (TA Institutional Deal) Swansea University 2025-08-18T13:46:46.7654816 2025-07-28T11:13:44.4778593 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Arslan Akbar 0009-0007-7102-8229 1 Suya Han 0009-0003-9075-8818 2 Naveed Urr Rehman 0009-0008-5704-9071 3 Kanwal Ahmed 4 Hassan Eshkiki 0000-0001-7795-453X 5 Fabio Caraffini 0000-0001-9199-7368 6 70043__34964__914a4dd4307146fb82f38e57c45590ee.pdf 70043.VoR.pdf 2025-08-18T13:43:09.9931560 Output 5234819 application/pdf Version of Record true © The Author(s) 2025. This article is licensed under a Creative Commons Attribution 4.0 International License. true eng http://creativecommons.org/licenses/by/4.0/
title Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
spellingShingle Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
Hassan Eshkiki
Fabio Caraffini
title_short Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
title_full Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
title_fullStr Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
title_full_unstemmed Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
title_sort Explainable breast cancer prediction from 3-dimensional dynamic contrast-enhanced magnetic resonance imaging
author_id_str_mv c9972b26a83de11ffe211070f26fe16b
d0b8d4e63d512d4d67a02a23dd20dfdb
author_id_fullname_str_mv c9972b26a83de11ffe211070f26fe16b_***_Hassan Eshkiki
d0b8d4e63d512d4d67a02a23dd20dfdb_***_Fabio Caraffini
author Hassan Eshkiki
Fabio Caraffini
author2 Arslan Akbar
Suya Han
Naveed Urr Rehman
Kanwal Ahmed
Hassan Eshkiki
Fabio Caraffini
format Journal article
container_title Applied Intelligence
container_volume 55
container_issue 13
publishDate 2025
institution Swansea University
issn 0924-669X
1573-7497
doi_str_mv 10.1007/s10489-025-06803-9
publisher Springer Science and Business Media LLC
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Deep learning models have been instrumental in extracting critical indicators for breast cancer diagnosis - the prevalent malignancy among women worldwide - from baseline magnetic resonance imaging. However, many existing models do not fully leverage the rich spatial information available in the 3D structure of medical imaging data, potentially overlooking important contextual details. This develops an explainable deep learning framework for classifying breast cancer that leverages the complete 3D and provides classification results alongside visual explanations of the decision-making process. The preprocessing pipeline is fed with 3D sequences containing ‘tumour’ and ‘non-tumour’ regions. It includes a 3D Adaptive Unsharp Mask (AUM) filter to reduce noise and augment image class, followed by normalisation and data augmentation. Classification is then achieved by training an augmented ResNet150 model. Three explainable artificial intelligence (XAI) techniques, including Shapley Additive Explanations, 3D Gradient-Weighted Class Activation Mapping, and Contextual Importance and Utility, are employed to provide improved interpretability. The model demonstrates state-of-the-art performance over the QIN-BREAST dataset, achieving testing accuracies of 98.861% for ‘tumours’ and 99.447% for ‘non-tumours’, as well as over the Duke Breast Cancer Dataset, where it achieves 99.104% for ‘tumours’ and 99.753% for ‘non-tumours’, while offering enhanced interpretability through XAI methods.
published_date 2025-08-01T05:25:34Z
_version_ 1851641282898690048
score 11.090009

Similar Items