{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T06:05:27Z","timestamp":1790661927354,"version":"4.1.0"},"reference-count":60,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"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. Genome Ed."],"abstract":"\n Background and Objective<\/jats:title>\n Accurate 3D segmentation of nasopharyngeal carcinoma (NPC) from MRI is critical for radiotherapy planning but remains challenging due to poor multi-center generalization, lack of spatial continuity, and high computational costs of global modeling. We aim to address these limitations by proposing a novel 3D framework.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We propose MambaGLA-NPC, a 3D deep learning framework integrating State Space Models (SSMs) and Convolutional Neural Networks (CNNs) in a U-shaped encoder-decoder architecture. A Conv-GLSSM module jointly extracts static local features via convolutions and global context via SSMs with a 3D local scanning strategy. A gated channel-spatial attention (GCS Att) module suppresses irrelevant noise in skip connections, and an NPC-specific adapter enables parameter-efficient fine-tuning across multi-center datasets.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n On a multi-center dataset of 1,098 NPC patients from three hospitals, our model achieves state-of-the-art performance on the internal test set (Dice: 80.47%, IoU: 67.83%, HD95: 4.09\u00a0mm). Under cross-center evaluation, it significantly outperforms prior methods (e.g., Dice of 87.13% and 76.70% on two external centers). The NPC-Adapter boosts cross-center Dice by up to 8.84% compared to no adaptation. Ablation studies confirm the contribution of each module, and the Conv-GLSSM reduces computational cost versus Transformers (21.66 GFLOPs vs. OOM for self-attention at\n \n \n \n \n \n 128<\/mml:mn>\n <\/mml:mrow>\n \n 3<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n resolution).\n <\/jats:p>\n <\/jats:sec>\n \n Conclusion<\/jats:title>\n MambaGLA-NPC is the first Mamba-based 3D NPC segmentation framework that effectively balances local-global feature integration with linear complexity, achieving superior accuracy and generalization across multi-center MRI data while maintaining high efficiency.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/fgeed.2026.1954329","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:50:17Z","timestamp":1790661017000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["MambaGLA-NPC: a 3D framework integrating CNN and state space models for robust multi-center nasopharyngeal carcinoma segmentation"],"prefix":"10.3389","volume":"8","author":[{"given":"Wei","family":"Jin","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence for Traditional Chinese Medicine, Zhejiang Chinese Medical University","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaobo","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Radiology, Guangdong Provincial People\u2019s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University","place":["Guangzhou, Guangdong, China"]},{"name":"Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application","place":["Guangzhou, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weifeng","family":"Jin","sequence":"additional","affiliation":[{"name":"School of Pharmaceutical Sciences, Zhejiang Chinese Medical University","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hai","family":"Liao","sequence":"additional","affiliation":[{"name":"Department of Medical Imaging Center, Guangxi Medical University Cancer Hospital, Guangxi Clinical Research Center for Medical Imaging Construction, Guangxi Key Clinical Specialty (Medical Imaging Department), Dominant Cultivation Discipline of Guangxi Medical University Cancer Hospital (Medical Imaging Department)","place":["Nanning, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Basic Medical Sciences, Zhejiang Chinese Medical University","place":["Hangzhou, Zhejiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"B1","unstructured":"Deep Learning Using Rectified Linear Units (relu)\n \n \n \n Agarap\n A. F.\n \n \n \n 2018"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1802.06955","article-title":"Recurrent residual convolutional neural network based on u-net (R2u-net) for medical image segmentation","author":"Alom","year":"2018"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2310.02972","article-title":"Fully automatic segmentation of gross target volume and organs-at-risk for radiotherapy planning of nasopharyngeal carcinoma","author":"Astaraki","year":"2023"},{"key":"B4","unstructured":"A comprehensive survey of mamba architectures for medical image analysis: classification, segmentation, restoration and beyond\n \n \n \n Bansal\n S.\n \n \n Madisetty\n S.\n \n \n Rehman\n M. Z. U.\n \n \n Raghaw\n C. S.\n \n \n Duggal\n G.\n \n \n Kumar\n N.\n \n \n 10.48550\/arXiv.2410.02362\n \n 2024"},{"key":"B5","doi-asserted-by":"publisher","first-page":"102471","DOI":"10.1016\/j.compmedimag.2024.102471","article-title":"Self-supervised learning on dual-sequence magnetic resonance imaging for automatic segmentation of nasopharyngeal carcinoma","volume":"118","author":"Cai","year":"2024","journal-title":"Comput. Med. Imaging Graph."},{"key":"B6","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1007\/978-981-96-4001-0_26","article-title":"Tun-gca: a novel approach for organ segmentation in nasopharyngeal carcinoma ct images","volume-title":"Human Brain and Artificial Intelligence (HBAI 2024)","author":"Che","year":"2025"},{"key":"B7","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1016\/S0140-6736(19)30956-0","article-title":"Nasopharyngeal carcinoma","volume":"394","author":"Chen","year":"2019","journal-title":"Lancet"},{"key":"B8","first-page":"424","article-title":"3d u-net: learning dense volumetric segmentation from sparse annotation","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)","author":"\u00c7i\u00e7ek","year":"2016"},{"key":"B9","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/j.neunet.2017.12.012","article-title":"Sigmoid-weighted linear units for neural network function approximation in reinforcement learning","volume":"107","author":"Elfwing","year":"2018","journal-title":"Neural Netw."},{"key":"B10","doi-asserted-by":"publisher","first-page":"107154","DOI":"10.1016\/j.bspc.2024.107154","article-title":"Pformer: an efficient cnn-transformer hybrid network with content-driven p-attention for 3d medical image segmentation","volume":"101","author":"Gao","year":"2025","journal-title":"Biomed. Signal Process. Control"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1822","DOI":"10.1109\/TMI.2018.2806309","article-title":"Automatic multi-organ segmentation on abdominal ct with dense v-networks","volume":"37","author":"Gibson","year":"2018","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.00752","article-title":"Mamba: Linear-Time Sequence Modeling with Selective State Spaces","author":"Gu","year":"2024"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2110.13985","article-title":"Combining Recurrent, Convolutional, and continuous-time Models with Linear state-space Layers","author":"Gu","year":"2021"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2111.00396","article-title":"Efficiently Modeling Long Sequences with Structured State Spaces","author":"Gu","year":"2022"},{"key":"B15","first-page":"272","article-title":"Swin unetr: swin transformers for semantic segmentation of brain tumors in mri images","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)","author":"Hatamizadeh","year":""},{"key":"B16","first-page":"574","article-title":"Unetr: transformers for 3d medical image segmentation","volume-title":"Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)","author":"Hatamizadeh","year":""},{"key":"B17","doi-asserted-by":"publisher","first-page":"472","DOI":"10.1007\/s10278-012-9520-4","article-title":"Region-based nasopharyngeal carcinoma lesion segmentation from mri using clustering-and classification-based methods with learning","volume":"26","author":"Huang","year":"2013","journal-title":"J. Digital Imaging"},{"key":"B18","doi-asserted-by":"publisher","first-page":"105869","DOI":"10.1016\/j.compbiomed.2022.105869","article-title":"Tg-net: combining transformer and gan for nasopharyngeal carcinoma tumor segmentation based on total-body uexplorer Pet\/Ct scanner","volume":"148","author":"Huang","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"B19","doi-asserted-by":"publisher","first-page":"565","DOI":"10.3390\/diagnostics10080565","article-title":"Multimodal brain tumor classification using deep learning and robust feature selection: a machine learning application for radiologists","volume":"10","author":"Khan","year":"2020","journal-title":"Diagnostics"},{"key":"B20","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","article-title":"Imagenet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun. ACM"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2209.15076","article-title":"3d ux-net: A Large Kernel Volumetric Convnet Modernizing Hierarchical Transformer for Medical Image Segmentation","author":"Lee","year":"2022"},{"key":"B22","doi-asserted-by":"publisher","first-page":"125518","DOI":"10.1016\/j.eswa.2024.125518","article-title":"Selective and multi-scale fusion mamba for medical image segmentation","volume":"261","author":"Li","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"B23","first-page":"2980","article-title":"Focal loss for dense object detection","volume-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV)","author":"Lin","year":"2017"},{"key":"B24","doi-asserted-by":"publisher","first-page":"12252","DOI":"10.1109\/ACCESS.2019.2892496","article-title":"Adaptive independent subspace analysis of brain magnetic resonance imaging data","volume":"7","author":"Lin","year":"2019","journal-title":"IEEE Access"},{"key":"B25","doi-asserted-by":"publisher","first-page":"6256","DOI":"10.1038\/s41598-020-63242-x","article-title":"Iouc3dsfcnn: segmentation of brain tumors via iou constraint 3d symmetric full convolution network with multimodal auto-context","volume":"10","author":"Liu","year":"2020","journal-title":"Sci. Rep."},{"key":"B26","doi-asserted-by":"crossref","DOI":"10.52202\/079017-3273","article-title":"Vmamba: visual state space model","author":"Liu","year":"2024"},{"key":"B27","article-title":"Decoupled weight decay regularization","volume-title":"International Conference on Learning Representations (ICLR)","author":"Loshchilov","year":"2019"},{"key":"B28","doi-asserted-by":"publisher","first-page":"1639","DOI":"10.1109\/TMI.2022.3144274","article-title":"Npcnet: jointly segment primary nasopharyngeal carcinoma tumors and metastatic lymph nodes in mr images","volume":"41","author":"Luo","year":"2022","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B29","first-page":"116","article-title":"Shufflenet v2: practical guidelines for efficient cnn architecture design","volume-title":"Proceedings of the European Conference on Computer Vision (ECCV)","author":"Ma","year":"2018"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.04722","article-title":"U-mamba: enhancing long-range dependency for biomedical image segmentation","author":"Ma","year":"2024"},{"key":"B31","doi-asserted-by":"crossref","first-page":"565","DOI":"10.1109\/3DV.2016.79","article-title":"V-net: fully convolutional neural networks for volumetric medical image segmentation","volume-title":"2016 Fourth International Conference on 3D Vision (3DV)","author":"Milletari","year":"2016"},{"key":"B32","doi-asserted-by":"publisher","first-page":"102525","DOI":"10.1016\/j.compmedimag.2025.102525","article-title":"Cqenet: a segmentation model for nasopharyngeal carcinoma based on confidence quantitative evaluation","volume":"123","author":"Qi","year":"2025","journal-title":"Comput. Med. Imaging Graph."},{"key":"B33","first-page":"234","article-title":"U-net: convolutional networks for biomedical image segmentation","volume-title":"International Conference on Medical Image Computing and computer-assisted Intervention","author":"Ronneberger","year":"2015"},{"key":"B34","first-page":"405","article-title":"Mednext: transformer-driven scaling of convnets for medical image segmentation","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)","author":"Roy","year":"2023"},{"key":"B35","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1186\/s12880-015-0068-x","article-title":"Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool","volume":"15","author":"Taha","year":"2015","journal-title":"BMC Med. Imaging"},{"key":"B36","doi-asserted-by":"publisher","first-page":"395","DOI":"10.1002\/cac2.12082","article-title":"Emerging radiotherapy technologies and trends in nasopharyngeal cancer","volume":"40","author":"Tseng","year":"2020","journal-title":"Cancer Commun."},{"key":"B37","first-page":"6000","article-title":"Attention is all you need","volume-title":"Advances in Neural Information Processing Systems (Neurips)","author":"Vaswani","year":"2017"},{"key":"B38","doi-asserted-by":"crossref","first-page":"1204","DOI":"10.1109\/ICPECA56706.2023.10075760","article-title":"A review on 3d convolutional neural network","volume-title":"2023 IEEE 3rd International Conference on Power, Electronics and Computer Applications (ICPECA)","author":"Wang","year":"2023"},{"key":"B39","first-page":"51","article-title":"Sam-med3d: towards general-purpose segmentation models for volumetric medical images","volume-title":"European Conference on Computer Vision (ECCV)","author":"Wang","year":"2024"},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2402.05079","article-title":"Mamba-Unet: Unet-Like Pure Visual Mamba for Medical Image Segmentation","author":"Wang","year":"2024"},{"key":"B41","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1007\/s11263-026-02864-6","article-title":"Parameter-efficient fine-tuning for pre-trained vision models: a survey and benchmark","volume":"134","author":"Xin","year":"2026","journal-title":"Int. J. Comput. Vis."},{"key":"B42","first-page":"578","article-title":"Segmamba: long-Range sequential modeling mamba for 3d medical image segmentation","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)","author":"Xing","year":"2024"},{"key":"B43","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1109\/TMI.2025.3589797","article-title":"Segmamba-v2: long-Range sequential modeling mamba for general 3d medical image segmentation","volume":"45","author":"Xing","year":"2025","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B44","doi-asserted-by":"publisher","first-page":"827991","DOI":"10.3389\/fonc.2022.827991","article-title":"Multiscale local enhancement deep convolutional networks for the automated 3d segmentation of gross tumor volumes in nasopharyngeal carcinoma: a multi-institutional dataset study","volume":"12","author":"Yang","year":"2022","journal-title":"Front. Oncol."},{"key":"B45","doi-asserted-by":"publisher","first-page":"1377366","DOI":"10.3389\/fonc.2024.1377366","article-title":"A transformer-based multi-task deep learning model for simultaneous t-stage identification and segmentation of nasopharyngeal carcinoma","volume":"14","author":"Yang","year":"2024","journal-title":"Front. Oncol."},{"key":"B46","doi-asserted-by":"publisher","first-page":"1051","DOI":"10.3390\/bioengineering12101051","article-title":"Lg-unet based segmentation and survival prediction of nasopharyngeal carcinoma using multimodal mri imaging","volume":"12","author":"Yang","year":"2025","journal-title":"Bioengineering"},{"key":"B47","article-title":"Pmamba: marrying perona malik diffusion with mamba for efficient pediatric echocardiographic left ventricular segmentation","author":"Ye","year":"2024"},{"key":"B48","doi-asserted-by":"publisher","first-page":"368504241232537","DOI":"10.1177\/00368504241232537","article-title":"Wet-unet: wavelet integrated efficient transformer networks for nasopharyngeal carcinoma tumor segmentation","volume":"107","author":"Zeng","year":"2024","journal-title":"Sci. Prog."},{"key":"B49","first-page":"6848","article-title":"Shufflenet: an extremely efficient convolutional neural network for Mobile devices","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Zhang","year":"2018"},{"key":"B50","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40644-019-0276-7","article-title":"The effects of volume of interest delineation on mri-based radiomics analysis: evaluation with two disease groups","volume":"19","author":"Zhang","year":"2019","journal-title":"Cancer Imaging"},{"key":"B51","doi-asserted-by":"publisher","first-page":"2531","DOI":"10.1109\/TMI.2020.2973595","article-title":"Generalizing deep learning for medical image segmentation to unseen domains via deep stacked transformation","volume":"39","author":"Zhang","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B52","doi-asserted-by":"publisher","first-page":"816672","DOI":"10.3389\/fonc.2021.816672","article-title":"Attr2u-net: a fully automated model for mri nasopharyngeal carcinoma segmentation based on spatial attention and residual recurrent convolution","volume":"11","author":"Zhang","year":"2022","journal-title":"Front. Oncol."},{"key":"B53","first-page":"1","article-title":"Lwv-unet: lightweight vision transformer unet for nasopharyngeal carcinoma image segmentation","volume-title":"2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI)","author":"Zhang","year":"2025"},{"key":"B54","doi-asserted-by":"publisher","first-page":"5447","DOI":"10.1109\/JBHI.2024.3405993","article-title":"Mmca-net: a multimodal cross attention transformer network for nasopharyngeal carcinoma tumor segmentation based on a total-body Pet\/Ct system","volume":"28","author":"Zhao","year":"2024","journal-title":"IEEE J. Biomed. Health Inf."},{"key":"B55","doi-asserted-by":"publisher","first-page":"108611","DOI":"10.1016\/j.cmpb.2025.108611","article-title":"Pmfsnet: polarized multi-scale feature self-attention network for lightweight medical image segmentation","volume":"261","author":"Zhong","year":"2025","journal-title":"Comput. Methods Programs Biomed."},{"key":"B56","first-page":"1364","article-title":"Nasopharyngeal carcinoma lesion segmentation from mr images by support vector machine","volume-title":"2006 3rd IEEE International Symposium on Biomedical Imaging (ISBI)","author":"Zhou","year":"2006"},{"key":"B57","first-page":"3","article-title":"Unet++: a nested u-net architecture for medical image segmentation","volume-title":"Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support (DLMIA\/ML-CDS)","author":"Zhou","year":"2018"},{"key":"B58","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2109.03201","article-title":"Nnformer: Interleaved Transformer for Volumetric Segmentation","author":"Zhou","year":"2021"},{"key":"B59","doi-asserted-by":"publisher","first-page":"126970","DOI":"10.1016\/j.neucom.2023.126970","article-title":"A multi-attention and depthwise separable convolution network for medical image segmentation","volume":"564","author":"Zhou","year":"2024","journal-title":"Neurocomputing"},{"key":"B60","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.09417","article-title":"Vision mamba: efficient visual representation learning with bidirectional state space model","author":"Zhu","year":"2024"}],"container-title":["Frontiers in Genome Editing"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fgeed.2026.1954329\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:50:24Z","timestamp":1790661024000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fgeed.2026.1954329\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":60,"alternative-id":["10.3389\/fgeed.2026.1954329"],"URL":"https:\/\/doi.org\/10.3389\/fgeed.2026.1954329","relation":{},"ISSN":["2673-3439"],"issn-type":[{"value":"2673-3439","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]},"article-number":"1954329"}}