{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T08:06:20Z","timestamp":1788941180798,"version":"build-2803163510"},"reference-count":21,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T00:00:00Z","timestamp":1788912000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurol."],"abstract":"<jats:p>Anatomical priors are often assumed to improve medical image segmentation, but their role can be ambiguous when strong image evidence is available. We study this question in a controlled small-sample dual-modal brain MRI setting with 13-class coarse brain-region segmentation from T1 and FLAIR images. Using a quality-controlled 78-case subset and a subject-disjoint five-fold protocol, we compare compact U-Net variants, coordinate channels, modality dropout, fold-wise index-space atlas conditioning, and a full-to-missing modality consistency variant, FW-AtlasMC. Fold-wise probabilistic spatial priors are built only from training labels, and an atlas-only baseline quantifies spatial prior strength without model training. Under full T1+FLAIR input, learned methods are tightly clustered in Dice, and atlas-informed models do not significantly improve full-modality Dice over a 2.5D U-Net baseline. However, among learned image-conditioned models under missing-modality inference, FW-AtlasMC improves FLAIR-only Dice and reduces average performance drop compared with 2.5D U-Net, modality dropout, and FW-Atlas. FW-AtlasMC combines atlas-informed inputs with missing-modality training and full-to-missing consistency. Validation-only sensitivity analysis further indicates that the chosen consistency weight represents a reasonable trade-off within the tested range. These results suggest that anatomical priors are better understood as structural stabilizers under degraded modality information, rather than as universal full-modality accuracy boosters.<\/jats:p>","DOI":"10.3389\/fneur.2026.1908175","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T07:12:56Z","timestamp":1788937976000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["When do anatomical priors help? Missing-modality robustness in small-sample dual-modal brain MRI region segmentation"],"prefix":"10.3389","volume":"17","author":[{"given":"Yuyan","family":"Zhang","sequence":"first","affiliation":[{"name":"School of Computer and Communication Engineering, University of Science and Technology Beijing","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,9]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"234","DOI":"10.1007\/978-3-319-24574-4_28","article-title":"\u201cU-Net: convolutional networks for biomedical image segmentation\u201d","author":"Ronneberger","year":"2015","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B2","doi-asserted-by":"publisher","first-page":"424","DOI":"10.1007\/978-3-319-46723-8_49","article-title":"\u201c3D U-Net: learning dense volumetric segmentation from sparse Annotation\u201d","author":"\u00c7i\u00e7ek","year":"2016","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B3","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1016\/j.media.2016.10.004","article-title":"Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation","volume":"36","author":"Kamnitsas","year":"2017","journal-title":"Med Image Anal"},{"key":"B4","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"B5","doi-asserted-by":"publisher","first-page":"839","DOI":"10.1016\/j.neuroimage.2005.02.018","article-title":"Unified segmentation","volume":"26","author":"Ashburner","year":"2005","journal-title":"Neuroimage"},{"key":"B6","doi-asserted-by":"publisher","first-page":"897","DOI":"10.1109\/42.811270","article-title":"Automated model-based tissue classification of MR images of the brain","volume":"18","author":"Van Leemput","year":"1999","journal-title":"IEEE Trans Med Imaging"},{"key":"B7","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1016\/j.neuroimage.2009.06.060","article-title":"Accurate and robust brain image alignment using boundary-based registration","volume":"48","author":"Greve","year":"2009","journal-title":"Neuroimage"},{"key":"B8","doi-asserted-by":"publisher","first-page":"469","DOI":"10.1007\/978-3-319-46723-8_54","article-title":"\u201cHeMIS: hetero-modal image segmentation\u201d","author":"Havaei","year":"2016","journal-title":"Medical Image Computing and Computer-Assisted Intervention Workshops"},{"key":"B9","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1007\/978-3-030-32245-8_9","article-title":"\u201cHetero-modal variational encoder-decoder for joint modality completion and segmentation\u201d","author":"Dorent","year":"2019","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1692","DOI":"10.1109\/TPAMI.2015.2461544","article-title":"ModDrop: adaptive multi-modal gesture recognition","volume":"38","author":"Neverova","year":"2016","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B11","doi-asserted-by":"crossref","unstructured":"\u201cModDrop++: a dynamic filter network with intra-subject co-training for multiple sclerosis lesion segmentation with missing modalities\u201d\n          \n          444\n          53\n          \n            \n              Charton\n              M\n            \n            \n              Gessert\n              N\n            \n            \n              Hepp\n              T\n            \n            \n              Appenzeller\n              S\n            \n            \n              Teuber\n              J\n            \n            \n              Kleesiek\n              J\n            \n          \n          \n            \n              Wang\n              L\n            \n            \n              Dou\n              Q\n            \n            \n              Fletcher\n              PT\n            \n            \n              Speidel\n              S\n            \n            \n              Li\n              S\n            \n          \n          Cham\n          Springer\n          Medical Image Computing and Computer-Assisted Intervention (MICCAI)\n          \n          2022","DOI":"10.1007\/978-3-031-16443-9_43"},{"key":"B12","first-page":"3975","article-title":"\u201cRFNET: region-aware fusion network for incomplete multi-modal brain tumor segmentation","author":"Ding","year":"2021"},{"key":"B13","doi-asserted-by":"publisher","first-page":"410","DOI":"10.1007\/978-3-030-87234-2_39","article-title":"\u201cACN: adversarial co-training network for brain tumor segmentation with missing modalities\u201d","author":"Wang","year":"2021","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B14","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1007\/978-3-031-16443-9_11","article-title":"\u201cmmFormer: multimodal medical transformer for incomplete multimodal learning of brain tumor segmentation\u201d","author":"Zhang","year":"2022","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B15","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1007\/978-3-031-16443-9_18","article-title":"\u201cModality-adaptive feature interaction for brain tumor segmentation with missing modalities","author":"Zhao","year":"2022","journal-title":"Medical Image Computing and Computer-Assisted Intervention (MICCAI)"},{"key":"B16","unstructured":"SMU-Net: style matching U-Net for brain tumor segmentation with missing modalities\n          \n          \n            \n              Azad\n              R\n            \n            \n              Khosravi\n              N\n            \n            \n              Merhof\n              D\n            \n          \n          arXiv:2204\n          \n          2022"},{"key":"B17","doi-asserted-by":"publisher","first-page":"108525","DOI":"10.1016\/j.neunet.2025.108525","article-title":"MGML: a plug-and-play meta-guided multi-modal learning framework for incomplete multimodal brain tumor segmentation","volume":"197","author":"Zou","year":"2026","journal-title":"Neural Netw"},{"key":"B18","unstructured":"CNN-based segmentation of medical imaging data. In arXiv [Preprint]\n          \n          \n            \n              Kayalibay\n              B\n            \n            \n              Jensen\n              G\n            \n            \n              van der Smagt\n              P\n            \n          \n          arXiv:1701\n          \n          2017"},{"key":"B19","doi-asserted-by":"publisher","first-page":"2663","DOI":"10.1109\/TMI.2018.2845918","article-title":"H-DenseUNet: hybrid densely connected UNet for liver and tumor segmentation from CT volumes","volume":"37","author":"Li","year":"2018","journal-title":"IEEE Trans Med Imaging."},{"key":"B20","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1038\/s41592-023-02151-z","article-title":"Metrics reloaded: recommendations for image analysis validation","volume":"21","author":"Maier-Hein","year":"2024","journal-title":"Nat Methods"},{"key":"B21","unstructured":"2025"}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1908175\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T07:12:57Z","timestamp":1788937977000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1908175\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,9]]},"references-count":21,"alternative-id":["10.3389\/fneur.2026.1908175"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2026.1908175","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,9]]},"article-number":"1908175"}}