{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T11:49:21Z","timestamp":1782301761057,"version":"3.54.5"},"reference-count":33,"publisher":"IOP Publishing","issue":"12","license":[{"start":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T00:00:00Z","timestamp":1782259200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T00:00:00Z","timestamp":1782259200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2026,6,28]]},"abstract":"Abstract<\/jats:title>\n \n Accurate glioma MRI segmentation is critical for clinical decision-making, yet voxel-wise annotation of volumetric MR images is expensive and time-consuming. We aim to maximize segmentation quality while reducing annotation cost through selective labeling. We develop a tumor-focused active learning AL framework that mirrors real-world annotation workflows and combines three complementary techniques: (i) a patient-specific, tumor-focused 40-slice window to preserve context while reducing computational requirements; (ii) a\n diverse<\/jats:italic>\n uncertainty committee spanning softmax entropy, margin, least confidence, variation ratio, and Bayesian AL by disagreement (BALD), where variation ratio and BALD are estimated via lightweight Monte Carlo dropout to capture epistemic uncertainty; and (iii)\n multi-criteria rank-level fusion<\/jats:italic>\n via Borda count to robustly aggregate uncertainty signals. In particular, we compute uncertainty both at the\n slice<\/jats:italic>\n level and at the\n region-of-interest (ROI)<\/jats:italic>\n level by summarizing pixel-wise uncertainty within tumor ROIs, then fuse all techniques and granularities with Borda rank to form a single acquisition order. A segmentation network is trained on an initial small labeled subset and then iteratively selects additional slices for annotation. Evaluation on UCSF\u2013PDGM and BraTS 2019 under realistic annotation budgets shows strong label efficiency and robustness: on UCSF\u2013PDGM, near-full performance is achieved with only\n \n \n \n <\/jats:tex-math>\n \n \n \u2248<\/mml:mo>\n 16<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n of the training data; on BraTS 2019, using\n \n \n \n <\/jats:tex-math>\n \n \n \u2248<\/mml:mo>\n 32<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n of the data yields performance comparable to models trained on all data, even surpassing the full-data model. Overall, the proposed framework combines the tumor-focused window with MC-dropout-enhanced,\n ROI-aware<\/jats:italic>\n uncertainty and Borda fusion to achieve accurate segmentation while aligning with real-world curation workflows.\n <\/jats:p>","DOI":"10.1088\/1361-6560\/ae7bac","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T22:55:34Z","timestamp":1781132134000},"page":"125020","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["ROI-aware uncertainty fusion for label-efficient glioma MRI segmentation"],"prefix":"10.1088","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-9224-8744","authenticated-orcid":true,"given":"Saher","family":"Mohamed","sequence":"first","affiliation":[{"name":"Nile University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7332-0759","authenticated-orcid":false,"given":"Ghada","family":"Khoriba","sequence":"additional","affiliation":[{"name":"Nile University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6571-9807","authenticated-orcid":true,"given":"Essam A","family":"Rashed","sequence":"additional","affiliation":[{"name":"University of Hyogo"},{"name":"University of Hyogo"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4601-7249","authenticated-orcid":false,"given":"Alaa","family":"Tharwat","sequence":"additional","affiliation":[{"name":"Bielefeld University of Applied Sciences and Arts"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,6,24]]},"reference":[{"key":"pmbae7bacbib1","article-title":"Deep batch active learning by diverse, uncertain gradient lower bounds","author":"Ash","year":"2020","type":"preprint"},{"key":"pmbae7bacbib2","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102531","type":"journal-article","article-title":"Estimating medical image registration error and confidence: a taxonomy and scoping review","volume":"81","author":"Bierbrier","year":"2022","journal-title":"Med. Image Anal."},{"key":"pmbae7bacbib3","article-title":"Multimodal Brain Tumor Segmentation Challenge CBICA Image Processing Portal University of Pennsylvania","author":"BraTS","year":"2019","type":"journal-article"},{"key":"pmbae7bacbib4","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1186\/s13244-023-01487-6","type":"journal-article","article-title":"An active learning approach to train a deep learning algorithm for tumor segmentation from brain MR images","volume":"14","author":"Boehringer","year":"2023","journal-title":"Insights Imaging"},{"key":"pmbae7bacbib5","article-title":"A survey on active learning and human-in-the-loop deep learning for medical image analysis","author":"Budd","year":"2019","type":"preprint"},{"key":"pmbae7bacbib6","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.220058","type":"journal-article","article-title":"The university of california san francisco preoperative diffuse glioma MRI (UCSF-PDGM) dataset","volume":"4","author":"Calabrese","year":"2022","journal-title":"Radiol. Artif. Intell."},{"key":"pmbae7bacbib7","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.220058","type":"journal-article","article-title":"The University of California San Francisco Preoperative Diffuse Glioma MRI The Cancer Imaging Archive","author":"Calabrese","year":"2022"},{"key":"pmbae7bacbib8","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222217","type":"journal-article","article-title":"Quantifying uncertainty in deep learning of radiologic images","volume":"308","author":"Faghani","year":"2023","journal-title":"Radiology"},{"key":"pmbae7bacbib9","doi-asserted-by":"crossref","DOI":"10.1016\/j.media.2023.102958","type":"preprint","article-title":"Active learning for medical image segmentation with stochastic batches","author":"Gaillochet","year":"2023"},{"key":"pmbae7bacbib10","article-title":"Dropout as a Bayesian approximation: representing model uncertainty in deep learning","author":"Gal","year":"2016","type":"preprint"},{"key":"pmbae7bacbib11","article-title":"Deep bayesian active learning with image data","author":"Gal","year":"2017","type":"preprint"},{"key":"pmbae7bacbib12","first-page":"pp 770","type":"conference-proceedings","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"pmbae7bacbib13","first-page":"pp 4700","type":"conference-proceedings","article-title":"Densely connected convolutional networks","author":"Huang","year":"2017"},{"key":"pmbae7bacbib14","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103223","type":"journal-article","article-title":"A review of uncertainty quantification in medical image analysis: probabilistic and non-probabilistic methods","volume":"97","author":"Huang","year":"2024","journal-title":"Med. Image Anal."},{"key":"pmbae7bacbib15","doi-asserted-by":"publisher","first-page":"pp 682","DOI":"10.1007\/978-3-030-00928-1_77","type":"conference-proceedings","article-title":"On the effect of inter-observer variability for a reliable estimation of uncertainty of medical image segmentation","author":"Jungo","year":"2018"},{"key":"pmbae7bacbib16","doi-asserted-by":"publisher","first-page":"2099","DOI":"10.1007\/s10278-024-01037-6","type":"journal-article","article-title":"Active learning in brain tumor segmentation with uncertainty sampling and annotation redundancy restriction","volume":"37","author":"Kim","year":"2024","journal-title":"J. Imaging Inform. Med."},{"key":"pmbae7bacbib17","article-title":"Batchbald: efficient and diverse batch acquisition for deep Bayesian active learning","author":"Kirsch","year":"2019","type":"other"},{"key":"pmbae7bacbib18","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP48485.2024.10446026","type":"preprint","article-title":"Breaking the barrier: selective uncertainty-based active learning for medical image segmentation","author":"Ma","year":"2024"},{"key":"pmbae7bacbib19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.59275\/j.melba.2022-354b","type":"journal-article","article-title":"Qu-brats: MICCAI brats 2020 challenge on quantifying uncertainty in brain tumor segmentation \u2013 analysis of ranking scores and benchmarking results","volume":"1","author":"Mehta","year":"2022","journal-title":"J. Mach. Learn. Biomed. Imaging"},{"key":"pmbae7bacbib20","doi-asserted-by":"publisher","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","type":"journal-article","article-title":"The multimodal brain tumor image segmentation benchmark (BRATS)","volume":"34","author":"Menze","year":"2015","journal-title":"IEEE Trans. Med. Imaging"},{"key":"pmbae7bacbib21","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1186\/s13104-022-06096-y","type":"journal-article","article-title":"Towards a guideline for evaluation metrics in medical image segmentation","volume":"15","author":"M\u00fcller","year":"2022","journal-title":"BMC Res. Notes"},{"key":"pmbae7bacbib22","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-24574-4_28","type":"preprint","article-title":"U-net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"pmbae7bacbib23","first-page":"pp 4510","type":"conference-proceedings","article-title":"Mobilenetv2: inverted residuals and linear bottlenecks","author":"Sandler","year":"2018"},{"key":"pmbae7bacbib24","article-title":"Active learning for convolutional neural networks: a core-set approach","author":"Sener","year":"2018","type":"preprint"},{"key":"pmbae7bacbib25","first-page":"pp 97","type":"conference-proceedings","article-title":"Image segmentation metrics in skin lesion: accuracy, sensitivity, specificity, dice coefficient, Jaccard index and Matthews correlation coefficient","author":"Setiawan","year":"2020"},{"key":"pmbae7bacbib26","article-title":"Active learning literature survey","author":"Settles","year":"2009","type":"other"},{"key":"pmbae7bacbib27","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","type":"preprint"},{"key":"pmbae7bacbib28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00546","type":"conference-proceedings","article-title":"Vaal: variational adversarial active learning","author":"Sinha","year":"2019"},{"key":"pmbae7bacbib29","first-page":"pp 6105","type":"conference-proceedings","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"key":"pmbae7bacbib30","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1016\/j.neucom.2019.01.103","type":"journal-article","article-title":"Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks","volume":"338","author":"Wang","year":"2019","journal-title":"Neurocomputing"},{"key":"pmbae7bacbib31","article-title":"Multi-faceted consistency learning with active cross-annotation for barely supervised 3D brain tumor segmentation","author":"Wu","year":"2024","type":"other"},{"key":"pmbae7bacbib32","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2019.00018","type":"other","article-title":"Learning loss for active learning","author":"Yoo","year":"2019"},{"key":"pmbae7bacbib33","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-72390-2_27","type":"other","article-title":"SBC-AL: structure and boundary consistency-based active learning for medical image segmentation","author":"Zhou","year":"2024"}],"container-title":["Physics in Medicine & Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T11:03:45Z","timestamp":1782299025000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae7bac"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,24]]},"references-count":33,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2026,6,24]]},"published-print":{"date-parts":[[2026,6,28]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/ae7bac","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,24]]},"assertion":[{"value":"ROI-aware uncertainty fusion for label-efficient glioma MRI segmentation","name":"article_title","label":"Article Title"},{"value":"Physics in Medicine & Biology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 Institute of Physics and Engineering in Medicine. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-01-05","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-06-10","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-06-24","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}