Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification

Wireless Capsule Endoscopy (WCE) image classification using deep learning models is hindered by data scarcity and model uncertainty. Labelling medical images is costly and time-consuming, limiting the availability of labelled data for training. To address these challenges, this work proposes ACT-WIS...

Full description

Saved in:
Bibliographic Details
Main Authors: Prabhanantha Kumar Muruganantham, Senthil Murugan Balakrishnan
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:Results in Engineering
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2590123025022467
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849318310460522496
author Prabhanantha Kumar Muruganantham
Senthil Murugan Balakrishnan
author_facet Prabhanantha Kumar Muruganantham
Senthil Murugan Balakrishnan
author_sort Prabhanantha Kumar Muruganantham
collection DOAJ
description Wireless Capsule Endoscopy (WCE) image classification using deep learning models is hindered by data scarcity and model uncertainty. Labelling medical images is costly and time-consuming, limiting the availability of labelled data for training. To address these challenges, this work proposes ACT-WISE, an active inter-consistency-driven semi-supervised learning framework that integrates Active Learning (AL) with Deep Learning (DL). ACT-WISE uses a teacher-student training methodology in which the model improves consistency by learning structural and semantic correlations from perturbations of unlabelled images. To reduce model uncertainty, batch acquisition selects the most informative samples from an unlabelled data pool based on minimum redundancy and maximum predictive entropy. Unlike traditional semi-supervised methods, ACT-WISE dynamically refines its selection strategy, optimising both label efficiency and model reliability. Based on the experimental results on the Kvasir-Capsule dataset for WCE image classification, the proposed ACT-WISE model demonstrates superior performance by achieving a classification accuracy of 0.97 and an AUC of 0.95, outperforming prior approaches. In addition, this work uses Monte Carlo dropout for model uncertainty estimation and assesses calibration reliability using Expected Calibration Error (ECE), providing interpretable and trustworthy prediction confidence scores. These results demonstrate that, with little manual annotation, ACT-WISE can greatly improve automatic anomaly detection in capsule endoscopy imaging, hence enhancing diagnostic support.
format Article
id doaj-art-fa8e4aecb0684c9dbd676c7de867d5f6
institution Kabale University
issn 2590-1230
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series Results in Engineering
spelling doaj-art-fa8e4aecb0684c9dbd676c7de867d5f62025-08-20T03:50:53ZengElsevierResults in Engineering2590-12302025-09-012710617410.1016/j.rineng.2025.106174Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classificationPrabhanantha Kumar Muruganantham0Senthil Murugan Balakrishnan1School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu 632014, IndiaCorresponding author.; School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu 632014, IndiaWireless Capsule Endoscopy (WCE) image classification using deep learning models is hindered by data scarcity and model uncertainty. Labelling medical images is costly and time-consuming, limiting the availability of labelled data for training. To address these challenges, this work proposes ACT-WISE, an active inter-consistency-driven semi-supervised learning framework that integrates Active Learning (AL) with Deep Learning (DL). ACT-WISE uses a teacher-student training methodology in which the model improves consistency by learning structural and semantic correlations from perturbations of unlabelled images. To reduce model uncertainty, batch acquisition selects the most informative samples from an unlabelled data pool based on minimum redundancy and maximum predictive entropy. Unlike traditional semi-supervised methods, ACT-WISE dynamically refines its selection strategy, optimising both label efficiency and model reliability. Based on the experimental results on the Kvasir-Capsule dataset for WCE image classification, the proposed ACT-WISE model demonstrates superior performance by achieving a classification accuracy of 0.97 and an AUC of 0.95, outperforming prior approaches. In addition, this work uses Monte Carlo dropout for model uncertainty estimation and assesses calibration reliability using Expected Calibration Error (ECE), providing interpretable and trustworthy prediction confidence scores. These results demonstrate that, with little manual annotation, ACT-WISE can greatly improve automatic anomaly detection in capsule endoscopy imaging, hence enhancing diagnostic support.http://www.sciencedirect.com/science/article/pii/S2590123025022467Active semi-supervised learningCapsule endoscopyUncertainty-aware modelsTeacher-student consistency regularisation
spellingShingle Prabhanantha Kumar Muruganantham
Senthil Murugan Balakrishnan
Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
Results in Engineering
Active semi-supervised learning
Capsule endoscopy
Uncertainty-aware models
Teacher-student consistency regularisation
title Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
title_full Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
title_fullStr Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
title_full_unstemmed Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
title_short Uncertainty-driven active learning in a deep semi-supervised framework for WCE image classification
title_sort uncertainty driven active learning in a deep semi supervised framework for wce image classification
topic Active semi-supervised learning
Capsule endoscopy
Uncertainty-aware models
Teacher-student consistency regularisation
url http://www.sciencedirect.com/science/article/pii/S2590123025022467
work_keys_str_mv AT prabhananthakumarmuruganantham uncertaintydrivenactivelearninginadeepsemisupervisedframeworkforwceimageclassification
AT senthilmuruganbalakrishnan uncertaintydrivenactivelearninginadeepsemisupervisedframeworkforwceimageclassification