{"paper":{"title":"Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Patch-MoE Mamba addresses limitations in Mamba models by using hierarchical patch-ordered scanning and mixture-of-experts fusion for medical image segmentation.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Fu, Diego Adame, Dongchul Kim, Erik Enriquez, Fabian Vazquez, Haoteng Tang, Huimin Li, Jinghao Yang, Jose A. Nunez, Pengfei Gu","submitted_at":"2026-05-18T00:42:04Z","abstract_excerpt":"CNN- and Transformer-based architectures have achieved strong performance in medical image segmentation, but CNNs are limited in modeling long-range dependencies, while Transformers often suffer from quadratic computational and memory complexity. State space models, especially Mamba-based networks, offer an efficient alternative with linear sequence complexity. However, existing Mamba segmentation models still face two limitations: pixel-wise directional scanning can disrupt local 2D spatial structure, and simple summation-based fusion of scan directions cannot adapt well to diverse object siz"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on five public polyp segmentation benchmarks and the ISIC 2017/2018 skin lesion segmentation datasets demonstrate the effectiveness and generality of Patch-MoE Mamba.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the hierarchical patch-ordered scanning mechanism preserves local spatial neighborhoods while capturing multi-scale context better than standard pixel-wise directional scanning, as stated in the problem setup for existing Mamba models.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Patch-MoE Mamba introduces patch-ordered hierarchical scanning and an MoE-based directional fusion module to improve Mamba segmentation models on polyp and skin lesion datasets.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Patch-MoE Mamba addresses limitations in Mamba models by using hierarchical patch-ordered scanning and mixture-of-experts fusion for medical image segmentation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8b3f508bb047653c26ed4a38b77d5241aaf3faa17a588d00b4ad2b8cfd255774"},"source":{"id":"2605.17719","kind":"arxiv","version":1},"verdict":{"id":"57d14627-b902-4e16-9ddb-90276f9136c9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:51:16.685165Z","strongest_claim":"Experiments on five public polyp segmentation benchmarks and the ISIC 2017/2018 skin lesion segmentation datasets demonstrate the effectiveness and generality of Patch-MoE Mamba.","one_line_summary":"Patch-MoE Mamba introduces patch-ordered hierarchical scanning and an MoE-based directional fusion module to improve Mamba segmentation models on polyp and skin lesion datasets.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the hierarchical patch-ordered scanning mechanism preserves local spatial neighborhoods while capturing multi-scale context better than standard pixel-wise directional scanning, as stated in the problem setup for existing Mamba models.","pith_extraction_headline":"Patch-MoE Mamba addresses limitations in Mamba models by using hierarchical patch-ordered scanning and mixture-of-experts fusion for medical image segmentation."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.17719/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:19.448978Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T22:00:55.518633Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"shingle_duplication","ran_at":"2026-05-19T21:49:43.468287Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T21:49:43.297824Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.501701Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T21:21:57.622679Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T21:21:57.404666Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"feb6a47b3010e128398c37a8f48ad88baf7e1241f776796e07b6c530147b3d9a"},"references":{"count":22,"sample":[{"doi":"","year":2020,"title":"PraNet: Parallel reverse attention network for polyp segmentation,","work_id":"ab3216da-5abc-4f58-a8ad-01fd97a21091","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2015,"title":"Automated polyp detection in colonoscopy videos using shape and context information,","work_id":"6892f33d-fbe2-4bd8-9446-7950b434caf6","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Keep your friends close & enemies farther: Debiasing contrastive learning with spatial priors in 3D radiology images,","work_id":"59d7fe67-2056-4d7f-aaa2-d46b67a09915","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Sli2vol+: Segmenting 3D medical images based on an object estimation guided correspondence flow network,","work_id":"4a404591-f3ce-4ad1-a2b9-ba7807f75e58","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2015,"title":"U-Net: Convolutional networks for biomedical image segmentation,","work_id":"cacc2998-30e6-45bf-927e-a538a94f1860","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":22,"snapshot_sha256":"d95383f94c5924b01e728ff4792998c1f633917b83e9b868838cf8e022d40900","internal_anchors":2},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}