{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T18:04:21Z","timestamp":1784311461674,"version":"3.55.0"},"reference-count":44,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T00:00:00Z","timestamp":1784246400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Scientific Research Innovation Capability Support Project for Young Faculty","award":["0401260003"],"award-info":[{"award-number":["0401260003"]}]},{"name":"National Science Fund for Excellent Overseas Scholars","award":["0401260011"],"award-info":[{"award-number":["0401260011"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["82472098"],"award-info":[{"award-number":["82472098"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Tianjin Natural Science Foundation - Outstanding Youth Project","award":["24JCJQJC00250"],"award-info":[{"award-number":["24JCJQJC00250"]}]},{"name":"Major Science and Technology Special Projects and Engineering - Major Project of National Key Laboratories","award":["24ZXZSSS00510"],"award-info":[{"award-number":["24ZXZSSS00510"]}]},{"name":"Autonomous Project of Haihe Laboratory of Brain-Computer Interaction and Human-Machine Integration","award":["2025E7-0121"],"award-info":[{"award-number":["2025E7-0121"]}]}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>The accurate delineation of ischemic stroke lesions in magnetic resonance imaging (MRI) is impeded by heterogeneous lesion morphology and the computational expense of modeling global context in three\u2011dimensional data. In cerebral infarction assessment, diffusion\u2011weighted imaging, apparent diffusion coefficient, T2\u2011weighted imaging and T2star sequences (including susceptibility weighted image processing) each offer complementary information, yet existing fusion strategies often fail to adapt to missing modalities or capture long\u2011range dependencies efficiently. Here we present GPMNet, a lightweight convolutional framework that integrates an adaptive multimodal feature fusion module\u2014employing dynamic cross\u2011attention to spatially weight and merge signals from all four MRI sequences\u2014and a gated parallel state\u2011space module that models global voxel interactions in linear time via dual gated branches. We trained the network end-to-end on the ATLAS R2.0 dataset and our own dataset collected at HuanHu Hospital (Tianjin, China), labeled as HHD. The training used a combined Dice\u2013binary cross-entropy and TOPK10 loss, and the outputs were refined using ensemble inference and connected-domain filtering. GPMNet achieved Dice coefficients of 0.6604 and 0.7171 on the two cohorts respectively, achieving superior results compared to other state-of-the-art algorithms. Moreover, the Grad-CAM\u2013based interpretability analysis confirms that the model\u2019s attention corresponds to true ischemic areas across modalities, offering visual evidence of its diagnostic reliability and enhancing the transparency of the segmentation process. Our approach delivered rapid, high\u2011precision stroke segmentation and establishes a scalable paradigm for resource\u2011efficient clinical imaging applications.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001517","type":"journal-article","created":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T17:34:39Z","timestamp":1784309679000},"page":"e0001517","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Adaptive multimodal fusion via Gated Parallel Mamba architecture for ultra-high-precision stroke lesion segmentation in medical imaging"],"prefix":"10.1371","volume":"5","author":[{"given":"Runnan","family":"He","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leshui","family":"Dong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuang","family":"Xia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Cao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Youwei","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Meijun","family":"Pang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kuo","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9540-4865","authenticated-orcid":true,"given":"Xiuyun","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dong","family":"Ming","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mark","family":"Haacke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,17]]},"reference":[{"issue":"8","key":"pdig.0001517.ref001","doi-asserted-by":"crossref","DOI":"10.1161\/CIR.0000000000001209","article-title":"2024 heart disease and stroke statistics: a report of US and global data from the American Heart Association","volume":"149","author":"SS Martin","year":"2024","journal-title":"Circulation"},{"issue":"5","key":"pdig.0001517.ref002","doi-asserted-by":"crossref","first-page":"1329","DOI":"10.1161\/STROKEAHA.123.045154","article-title":"Moving from CT to MRI paradigm in acute ischemic stroke: feasibility, effects on stroke diagnosis and long-term outcomes","volume":"55","author":"CM Rapillo","year":"2024","journal-title":"Stroke"},{"issue":"7","key":"pdig.0001517.ref003","doi-asserted-by":"crossref","first-page":"3372","DOI":"10.1109\/JBHI.2023.3270861","article-title":"Simulated quantum mechanics-based joint learning network for stroke lesion segmentation and TICI grading","volume":"27","author":"L Liu","year":"2023","journal-title":"IEEE J Biomed Health Inform"},{"key":"pdig.0001517.ref004","doi-asserted-by":"crossref","first-page":"106717","DOI":"10.1016\/j.compbiomed.2023.106717","article-title":"SAN-Net: learning generalization to unseen sites for stroke lesion segmentation with self-adaptive normalization","volume":"156","author":"W Yu","year":"2023","journal-title":"Comput Biol Med"},{"key":"pdig.0001517.ref005","doi-asserted-by":"crossref","first-page":"94871","DOI":"10.1109\/ACCESS.2020.2995632","article-title":"Automatic segmentation of stroke lesions in non-contrast computed tomography datasets with convolutional neural networks","volume":"8","author":"A Tuladhar","year":"2020","journal-title":"IEEE Access"},{"key":"pdig.0001517.ref006","doi-asserted-by":"crossref","first-page":"105941","DOI":"10.1016\/j.compbiomed.2022.105941","article-title":"Brain stroke classification and segmentation using encoder-decoder based deep convolutional neural networks","volume":"149","author":"S Yal\u00e7\u0131n","year":"2022","journal-title":"Comput Biol Med"},{"key":"pdig.0001517.ref007","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015: 18th International Conference; 2015 Oct 5\u20139; Munich, Germany, proceedings, part III 18. Springer International Publishing; 2015. p. 234\u201341.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"pdig.0001517.ref008","doi-asserted-by":"crossref","unstructured":"Wang X, Girshick R, Gupta A, et al. Non-local neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2018. p. 7794\u2013803.","DOI":"10.1109\/CVPR.2018.00813"},{"key":"pdig.0001517.ref009","article-title":"Attention is all you need","author":"A Vaswani","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"pdig.0001517.ref010","doi-asserted-by":"crossref","first-page":"108345","DOI":"10.1016\/j.compeleceng.2022.108345","article-title":"Multi-scale long-range interactive and regional attention network for stroke lesion segmentation","volume":"103","author":"Z Wu","year":"2022","journal-title":"Comput Electr Eng"},{"issue":"6","key":"pdig.0001517.ref011","doi-asserted-by":"crossref","first-page":"2303","DOI":"10.1109\/TMI.2024.3362879","article-title":"Hybrid CNN-transformer network with circular feature interaction for acute ischemic stroke lesion segmentation on non-contrast CT scans","volume":"43","author":"H Kuang","year":"2024","journal-title":"IEEE Trans Med Imaging"},{"key":"pdig.0001517.ref012","article-title":"Efficiently modeling long sequences with structured state spaces.","author":"A Gu","year":"2021"},{"key":"pdig.0001517.ref013","article-title":"Vmambair: visual state space model for image restoration.","author":"Y Shi","year":"2024"},{"key":"pdig.0001517.ref014","article-title":"Mamba: linear-time sequence modeling with selective state spaces.","author":"A Gu","year":"2023"},{"issue":"1","key":"pdig.0001517.ref015","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1007\/s12204-021-2273-9","article-title":"Application of deep learning method on ischemic stroke lesion segmentation","volume":"27","author":"Y Zhang","year":"2021","journal-title":"J Shanghai Jiaotong Univ (Sci)"},{"key":"pdig.0001517.ref016","article-title":"CNN-based segmentation of medical imaging data.","author":"B Kayalibay","year":"2017"},{"issue":"3","key":"pdig.0001517.ref017","doi-asserted-by":"crossref","first-page":"940","DOI":"10.1109\/TCBB.2019.2939522","article-title":"D-UNet: a dimension-fusion U shape network for chronic stroke lesion segmentation","volume":"18","author":"Y Zhou","year":"2021","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"key":"pdig.0001517.ref018","doi-asserted-by":"crossref","unstructured":"Qi K, Yang H, Li C, et al. X-net: Brain stroke lesion segmentation based on depthwise separable convolution and long-range dependencies. Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference; 2019 Oct 13\u201317; Shenzhen, China, Proceedings, Part III 22. Springer International Publishing; 2019. p. 247\u2013255.","DOI":"10.1007\/978-3-030-32248-9_28"},{"key":"pdig.0001517.ref019","doi-asserted-by":"crossref","first-page":"105524","DOI":"10.1016\/j.cmpb.2020.105524","article-title":"CSNet: a new DeepNet framework for ischemic stroke lesion segmentation","volume":"193","author":"A Kumar","year":"2020","journal-title":"Comput Methods Programs Biomed"},{"key":"pdig.0001517.ref020","article-title":"Fractalnet: ultra-deep neural networks without residuals.","author":"G Larsson","year":"2016"},{"issue":"11","key":"pdig.0001517.ref021","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1007\/s10462-024-10900-5","article-title":"Transformers-based architectures for stroke segmentation: a review","volume":"57","author":"Y Zafari-Ghadim","year":"2024","journal-title":"Artif Intell Rev"},{"key":"pdig.0001517.ref022","doi-asserted-by":"crossref","first-page":"120637","DOI":"10.1016\/j.eswa.2023.120637","article-title":"W-Net: a boundary-enhanced segmentation network for stroke lesions","volume":"230","author":"Z Wu","year":"2023","journal-title":"Expert Syst Appl"},{"issue":"9","key":"pdig.0001517.ref023","doi-asserted-by":"crossref","first-page":"340","DOI":"10.1049\/ell2.12444","article-title":"METrans: multi\u2010encoder transformer for ischemic stroke segmentation","volume":"58","author":"J Wang","year":"2022","journal-title":"Electron Lett"},{"key":"pdig.0001517.ref024","doi-asserted-by":"crossref","unstructured":"Luo C, Zhang J, Chen X, et al. UCATR: Based on CNN and transformer encoding and cross-attention decoding for lesion segmentation of acute ischemic stroke in non-contrast computed tomography images. 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE; 2021. p. 3565\u20138.","DOI":"10.1109\/EMBC46164.2021.9630336"},{"key":"pdig.0001517.ref025","doi-asserted-by":"crossref","unstructured":"Feng P, Ni B, Cai X, et al. Utransnet: transformer within u-net for stroke lesion segmentation. 2022 IEEE 25th International Conference on Computer Supported Cooperative Work in Design (CSCWD). IEEE; 2022. p. 359\u201364.","DOI":"10.1109\/CSCWD54268.2022.9776250"},{"key":"pdig.0001517.ref026","doi-asserted-by":"crossref","unstructured":"Wang G, Li W, Ourselin S, et al. Automatic brain tumor segmentation using cascaded anisotropic convolutional neural networks. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: Third International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017; 2017 Sept 14; Quebec City, QC, Canada, Revised Selected Papers 3. Springer International Publishing; 2018. p. 178\u201390.","DOI":"10.1007\/978-3-319-75238-9_16"},{"key":"pdig.0001517.ref027","doi-asserted-by":"crossref","unstructured":"Zhou C, Ding C, Lu Z, et al. One-pass multi-task convolutional neural networks for efficient brain tumor segmentation. Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2018: 21st International Conference; 2018 Sept 16\u201320; Granada, Spain, Proceedings, Part III 11. Springer International Publishing; 2018. p. 637\u201345.","DOI":"10.1007\/978-3-030-00931-1_73"},{"key":"pdig.0001517.ref028","article-title":"Medmamba: vision mamba for medical image classification.","author":"Y Yue","year":"2024"},{"key":"pdig.0001517.ref029","article-title":"P-mamba: marrying perona malik diffusion with mamba for efficient pediatric echocardiographic left ventricular segmentation.","author":"Z Ye","year":"2024"},{"issue":"7","key":"pdig.0001517.ref030","doi-asserted-by":"crossref","first-page":"629","DOI":"10.1109\/34.56205","article-title":"Scale-space and edge detection using anisotropic diffusion","volume":"12","author":"P Perona","year":"1990","journal-title":"IEEE Trans Pattern Anal Machine Intell"},{"key":"pdig.0001517.ref031","article-title":"Promamba: prompt-mamba for polyp segmentation.","author":"J Xie","year":"2024"},{"key":"pdig.0001517.ref032","article-title":"Lightm-unet: mamba assists in lightweight unet for medical image segmentation.","author":"W Liao","year":"2024"},{"key":"pdig.0001517.ref033","article-title":"U-mamba: enhancing long-range dependency for biomedical image segmentation.","author":"J Ma","year":"2024"},{"key":"pdig.0001517.ref034","doi-asserted-by":"crossref","unstructured":"Liu J, Yang H, Zhou H Y, et al. Swin-umamba: Mamba-based unet with imagenet-based pretraining. International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer Nature Switzerland; 2024. p. 615\u201325.","DOI":"10.1007\/978-3-031-72114-4_59"},{"key":"pdig.0001517.ref035","article-title":"Ultralight vm-unet: parallel vision mamba significantly reduces parameters for skin lesion segmentation.","author":"R Wu","year":"2024"},{"issue":"2","key":"pdig.0001517.ref036","doi-asserted-by":"crossref","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":"F Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"pdig.0001517.ref037","article-title":"Learning with average top-k loss","volume":"30","author":"Y Fan","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"pdig.0001517.ref038","article-title":"Efficiently modeling long sequences with structured state spaces.","author":"A Gu","year":"2021"},{"key":"pdig.0001517.ref039","unstructured":"Dauphin Y N, Fan A, Auli M, et al. Language modeling with gated convolutional networks. International Conference on Machine Learning. PMLR; 2017. p. 933\u201341."},{"issue":"1","key":"pdig.0001517.ref040","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1038\/s41597-022-01401-7","article-title":"A large, curated, open-source stroke neuroimaging dataset to improve lesion segmentation algorithms","volume":"9","author":"S-L Liew","year":"2022","journal-title":"Sci Data"},{"key":"pdig.0001517.ref041","article-title":"Transformer utilization in medical image segmentation networks.","author":"S Roy","year":"2023"},{"key":"pdig.0001517.ref042","doi-asserted-by":"crossref","unstructured":"Myronenko A. 3D MRI brain tumor segmentation using autoencoder regularization. Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 4th International Workshop, BrainLes 2018, Held in Conjunction with MICCAI 2018; 2018 Sept 16; Granada, Spain, Revised Selected Papers, Part II 4. Springer International Publishing; 2019. p. 311\u201320.","DOI":"10.1007\/978-3-030-11726-9_28"},{"key":"pdig.0001517.ref043","doi-asserted-by":"crossref","unstructured":"Hatamizadeh A, Nath V, Tang Y, et al. Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images. International MICCAI Brainlesion Workshop. Cham: Springer International Publishing; 2021. p. 272\u201384.","DOI":"10.1007\/978-3-031-08999-2_22"},{"key":"pdig.0001517.ref044","doi-asserted-by":"crossref","unstructured":"Kasula V K, Yenugula M, Konda B, et al. Federated learning with secure aggregation for privacy-preserving deep learning in IoT environments. 2025 IEEE Conference on Computer Applications (ICCA). IEEE; 2025. p. 1\u20137.","DOI":"10.1109\/ICCA65395.2025.11011120"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001517","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T17:34:46Z","timestamp":1784309686000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001517"}},"subtitle":[],"editor":[{"given":"Ismini","family":"Lourentzou","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,17]]},"references-count":44,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,17]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001517","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,17]]}}}