Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach

Deep Learning has contributed significantly to the advances made in the fields of Medical Imaging and Computer Aided Diagnosis (CAD). Although a variety of Deep Learning (DL) models exist for the purposes of image classification in the medical domain, more analysis needs to be conducted on their dec...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Varam, Dara (author)
مؤلفون آخرون: Mitra, Rohan (author), Mkadmi, Meriam (author), Riyas, Radi Aman (author), Abuhani, Diaa Addeen (author), Dhou, Salam (author), Alzaatreh, Ayman (author)
التنسيق: article
منشور في: 2023
الموضوعات:
الوصول للمادة أونلاين:https://hdl.handle.net/11073/33567
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1870676404103806976
author Varam, Dara
author2 Mitra, Rohan
Mkadmi, Meriam
Riyas, Radi Aman
Abuhani, Diaa Addeen
Dhou, Salam
Alzaatreh, Ayman
author2_role author
author
author
author
author
author
author_facet Varam, Dara
Mitra, Rohan
Mkadmi, Meriam
Riyas, Radi Aman
Abuhani, Diaa Addeen
Dhou, Salam
Alzaatreh, Ayman
author_role author
dc.creator.none.fl_str_mv Varam, Dara
Mitra, Rohan
Mkadmi, Meriam
Riyas, Radi Aman
Abuhani, Diaa Addeen
Dhou, Salam
Alzaatreh, Ayman
dc.date.none.fl_str_mv 2023-09-25
2026-06-29T05:10:43Z
2026-06-29T05:10:43Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv Varam, D., Mitra, R., Mkadmi, M., Riyas, R. A., Abuhani, D. A., Dhou, S., & Alzaatreh, A. (2023). Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach. IEEE Access, 11, 105262–105280. https://doi.org/10.1109/access.2023.3319068
2169-3536
https://hdl.handle.net/11073/33567
10.1109/access.2023.3319068
dc.language.none.fl_str_mv en
dc.publisher.none.fl_str_mv IEEE
dc.relation.none.fl_str_mv https://doi.org/10.1109/access.2023.3319068
dc.rights.none.fl_str_mv Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subject.none.fl_str_mv Deep learning
Explainable AI
Gastrointestinal diseases
Machine learning
Vision transformer
Wireless capsule endoscopy
dc.title.none.fl_str_mv Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
dc.type.none.fl_str_mv Peer-Reviewed
Published version
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description Deep Learning has contributed significantly to the advances made in the fields of Medical Imaging and Computer Aided Diagnosis (CAD). Although a variety of Deep Learning (DL) models exist for the purposes of image classification in the medical domain, more analysis needs to be conducted on their decision-making processes. For this reason, several novel Explainable AI (XAI) techniques have been proposed in recent years to better understand DL models. Currently, medical professionals rely on visual inspections to diagnose potential diseases in endoscopic imaging in the preliminary stages. However, we believe that the use of automated systems can enhance both the efficiency for such diagnoses. The aim of this study is to increase the reliability of model predictions within the field of endoscopic imaging by implementing several transfer learning models on a balanced subset of Kvasir-capsule, a Wireless Capsule Endoscopy imaging dataset. This subset includes the top 9 classes of the dataset for training and testing. The results obtained were an F1-score of 97% ±1% for the Vision Transformer model, although other models such as MobileNetv3Large and ResNet152v2 were also able to achieve F1-scores of over 90%. These are currently the highest-reported metrics on this data, improving upon prior studies done on the same dataset. The heatmaps of several XAI techniques, including GradCAM, GradCAM++, LayersCAM, LIME, and SHAP have been presented in image form and evaluated according to their highlighted regions of importance. This is in an effort to better understand the decisions of the top-performing DL models and look beyond their black-box nature.
format article
id aus_e16cbec14b29790ce50a82fa6b217855
identifier_str_mv Varam, D., Mitra, R., Mkadmi, M., Riyas, R. A., Abuhani, D. A., Dhou, S., & Alzaatreh, A. (2023). Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach. IEEE Access, 11, 105262–105280. https://doi.org/10.1109/access.2023.3319068
2169-3536
10.1109/access.2023.3319068
language_invalid_str_mv en
network_acronym_str aus
network_name_str aus
oai_identifier_str oai:repository.aus.edu:11073/33567
publishDate 2023
publisher.none.fl_str_mv IEEE
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv Attribution-NonCommercial-NoDerivatives 4.0 International
http://creativecommons.org/licenses/by-nc-nd/4.0/
spelling Wireless Capsule Endoscopy Image Classification: An Explainable AI ApproachVaram, DaraMitra, RohanMkadmi, MeriamRiyas, Radi AmanAbuhani, Diaa AddeenDhou, SalamAlzaatreh, AymanDeep learningExplainable AIGastrointestinal diseasesMachine learningVision transformerWireless capsule endoscopyDeep Learning has contributed significantly to the advances made in the fields of Medical Imaging and Computer Aided Diagnosis (CAD). Although a variety of Deep Learning (DL) models exist for the purposes of image classification in the medical domain, more analysis needs to be conducted on their decision-making processes. For this reason, several novel Explainable AI (XAI) techniques have been proposed in recent years to better understand DL models. Currently, medical professionals rely on visual inspections to diagnose potential diseases in endoscopic imaging in the preliminary stages. However, we believe that the use of automated systems can enhance both the efficiency for such diagnoses. The aim of this study is to increase the reliability of model predictions within the field of endoscopic imaging by implementing several transfer learning models on a balanced subset of Kvasir-capsule, a Wireless Capsule Endoscopy imaging dataset. This subset includes the top 9 classes of the dataset for training and testing. The results obtained were an F1-score of 97% ±1% for the Vision Transformer model, although other models such as MobileNetv3Large and ResNet152v2 were also able to achieve F1-scores of over 90%. These are currently the highest-reported metrics on this data, improving upon prior studies done on the same dataset. The heatmaps of several XAI techniques, including GradCAM, GradCAM++, LayersCAM, LIME, and SHAP have been presented in image form and evaluated according to their highlighted regions of importance. This is in an effort to better understand the decisions of the top-performing DL models and look beyond their black-box nature.IEEE2026-06-29T05:10:43Z2026-06-29T05:10:43Z2023-09-25Peer-ReviewedPublished versioninfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfVaram, D., Mitra, R., Mkadmi, M., Riyas, R. A., Abuhani, D. A., Dhou, S., & Alzaatreh, A. (2023). Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach. IEEE Access, 11, 105262–105280. https://doi.org/10.1109/access.2023.33190682169-3536https://hdl.handle.net/11073/3356710.1109/access.2023.3319068enhttps://doi.org/10.1109/access.2023.3319068Attribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/oai:repository.aus.edu:11073/335672026-06-29T06:13:14Z
spellingShingle Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
Varam, Dara
Deep learning
Explainable AI
Gastrointestinal diseases
Machine learning
Vision transformer
Wireless capsule endoscopy
status_str publishedVersion
title Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
title_full Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
title_fullStr Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
title_full_unstemmed Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
title_short Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
title_sort Wireless Capsule Endoscopy Image Classification: An Explainable AI Approach
topic Deep learning
Explainable AI
Gastrointestinal diseases
Machine learning
Vision transformer
Wireless capsule endoscopy
url https://hdl.handle.net/11073/33567