000286092 001__ 286092
000286092 005__ 20260413130622.0
000286092 0247_ $$2doi$$a10.1016/j.neucom.2026.133460
000286092 0247_ $$2ISSN$$a0925-2312
000286092 0247_ $$2ISSN$$a1872-8286
000286092 037__ $$aDZNE-2026-00388
000286092 082__ $$a610
000286092 1001_ $$00000-0001-7594-1188$$aChatterjee, Soumick$$b0
000286092 245__ $$aWeakly-supervised segmentation using inherently-explainable classification models and their application to brain tumour classification
000286092 260__ $$aAmsterdam$$bElsevier$$c2026
000286092 3367_ $$2DRIVER$$aarticle
000286092 3367_ $$2DataCite$$aOutput Types/Journal article
000286092 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1776078240_32195
000286092 3367_ $$2BibTeX$$aARTICLE
000286092 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000286092 3367_ $$00$$2EndNote$$aJournal Article
000286092 520__ $$aDeep learning has demonstrated significant potential in medical imaging; however, the opacity of “black-box” models hinders clinical trust, while segmentation tasks typically necessitate laborious, hard-to-obtain pixel-wise annotations. To address these challenges simultaneously, this paper introduces a framework for three inherently explainable classifiers (GP-UNet, GP-ShuffleUNet, and GP-ReconResNet). By integrating a global pooling mechanism, these networks generate localisation heatmaps that directly influence classification decisions, offering inherent interpretability without relying on potentially unreliable post-hoc methods. These heatmaps are subsequently thresholded to achieve weakly-supervised segmentation, requiring only image-level classification labels for training. Validated on two datasets for multi-class brain tumour classification, the proposed models achieved a peak F1-score of 0.93. For the weakly-supervised segmentation task, a median Dice score of 0.728 (95% CI: 0.715–0.739) was recorded. Notably, on a subset of tumour-only images, the best model achieved an accuracy of 98.7%, outperforming state-of-the-art glioma grading binary classifiers. Furthermore, comparative Precision-Recall analysis validated the framework’s robustness against severe class imbalance, establishing a direct correlation between diagnostic confidence and segmentation fidelity. These results demonstrate that the proposed framework successfully combines high diagnostic accuracy with essential transparency, offering a promising direction for trustworthy clinical decision support.
000286092 536__ $$0G:(DE-HGF)POF4-353$$a353 - Clinical and Health Care Research (POF4-353)$$cPOF4-353$$fPOF IV$$x0
000286092 588__ $$aDataset connected to CrossRef, Journals: pub.dzne.de
000286092 7001_ $$00000-0003-0987-505X$$aYassin, Hadya$$b1
000286092 7001_ $$aDubost, Florian$$b2
000286092 7001_ $$00000-0003-4311-0624$$aNürnberger, Andreas$$b3
000286092 7001_ $$0P:(DE-2719)2810706$$aSpeck, Oliver$$b4$$eLast author
000286092 773__ $$0PERI:(DE-600)1479006-3$$a10.1016/j.neucom.2026.133460$$gVol. 682, p. 133460 -$$p133460$$tNeurocomputing$$v682$$x0925-2312$$y2026
000286092 8564_ $$uhttps://pub.dzne.de/record/286092/files/DZNE-2026-00388_Restricted.pdf
000286092 8564_ $$uhttps://pub.dzne.de/record/286092/files/DZNE-2026-00388_Restricted.pdf?subformat=pdfa$$xpdfa
000286092 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)2810706$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b4$$kDZNE
000286092 9131_ $$0G:(DE-HGF)POF4-353$$1G:(DE-HGF)POF4-350$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lNeurodegenerative Diseases$$vClinical and Health Care Research$$x0
000286092 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bNEUROCOMPUTING : 2022$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2025-11-11
000286092 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bNEUROCOMPUTING : 2022$$d2025-11-11
000286092 9201_ $$0I:(DE-2719)1340009$$kAG Speck$$lLinking Imaging Projects$$x0
000286092 980__ $$ajournal
000286092 980__ $$aEDITORS
000286092 980__ $$aVDBINPRINT
000286092 980__ $$aI:(DE-2719)1340009
000286092 980__ $$aUNRESTRICTED