{"paper":{"title":"Virtual Nodes Guided Dynamic Graph Neural Network for Brain Tumor Segmentation with Missing Modalities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A graph neural network with virtual nodes and dynamic connections segments brain tumors effectively even with missing MRI modalities.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chao Yao, Jiao Pan, Sha Tao, Yu Guo","submitted_at":"2026-05-16T08:40:01Z","abstract_excerpt":"Multimodal magnetic resonance imaging (MRI) is crucial for brain tumor segmentation, with many methods leveraging its four key modalities to capture complementary information for effective sub-region analysis. However, the absence of several modalities is very common in practice, leading to severe performance degradation in existing full-modality segmentation methods. Limited by the structured data model, recent works often adopt a multi-stage training strategy for full-modality and missing-modality scenarios, which increases training costs and inadequately addresses the interference of miss. "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Extensive experiments on the BRATS-2018 and BRATS-2020 datasets demonstrate that our method outperforms the state-of-the-art methods on almost all subsets of incomplete modalities.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That dynamically adjusting the adjacency matrix based on modality availability will preserve beneficial information flow while mitigating interference effects caused by missing modalities without introducing new biases or artifacts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A one-stage graph framework with modality-specific virtual nodes and dynamic adjacency adjustment for robust brain tumor segmentation under arbitrary missing MRI modalities, outperforming SOTA on BRATS-2018 and BRATS-2020 incomplete subsets.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A graph neural network with virtual nodes and dynamic connections segments brain tumors effectively even with missing MRI modalities.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5c2c8db5b316c7c41d4b69998969da751b51e6629a723ca2fcafbdf7398f3d8e"},"source":{"id":"2605.16880","kind":"arxiv","version":1},"verdict":{"id":"a6acd833-508d-45a3-b79b-bec0fb9b38e2","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:48:43.918762Z","strongest_claim":"Extensive experiments on the BRATS-2018 and BRATS-2020 datasets demonstrate that our method outperforms the state-of-the-art methods on almost all subsets of incomplete modalities.","one_line_summary":"A one-stage graph framework with modality-specific virtual nodes and dynamic adjacency adjustment for robust brain tumor segmentation under arbitrary missing MRI modalities, outperforming SOTA on BRATS-2018 and BRATS-2020 incomplete subsets.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That dynamically adjusting the adjacency matrix based on modality availability will preserve beneficial information flow while mitigating interference effects caused by missing modalities without introducing new biases or artifacts.","pith_extraction_headline":"A graph neural network with virtual nodes and dynamic connections segments brain tumors effectively even with missing MRI modalities."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16880/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T21:01:23.104620Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.209840Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.291387Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.368623Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"9e5944b43f64452434c8542b1bc1b930c971ebbe48ab574e56e2a65e68d55b8f"},"references":{"count":39,"sample":[{"doi":"","year":2012,"title":"Slic superpix- els compared to state-of-the-art superpixel methods.IEEE transactions on pattern analysis and machine intelligence, 34(11):2274–2282, 2012","work_id":"744d37b9-ef7b-4c25-ba91-dfb7b5a5a49c","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Smu-net: Style matching u-net for brain tumor segmentation with missing modalities","work_id":"90118646-120c-46af-a2ce-71eaed1ac9fb","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Robust multimodal brain tumor seg- mentation via feature disentanglement and gated fusion","work_id":"7b3c5917-53f4-4801-92b4-c4ffce088ef8","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Learning with privileged multimodal knowledge for unimodal segmentation.IEEE transactions on medical imaging, 41(3):621–632, 2021","work_id":"56d1f87c-2eb0-4e5f-828d-3fe82e53d8c1","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Rfnet: Region-aware fusion network for incomplete multi-modal brain tumor seg- mentation","work_id":"fb3195dd-ee6f-443d-bb42-02592a2e9a25","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":39,"snapshot_sha256":"b193dc4459a4e1113acc5959f076af22b2a8d9df14d6e54535737bb08fec01fa","internal_anchors":4},"formal_canon":{"evidence_count":1,"snapshot_sha256":"0f2adf429bb4d132aef123434c801c67121dc31f7e8d6c9ce5cc16a6dbef9576"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}