{"url":"/sota/medical-image-segmentation-on-synapse-multi","task":{"name":"Medical Image Segmentation","url":"/task/medical-image-segmentation","note":null},"dataset":{"name":"Synapse multi-organ CT","url":"/dataset/miccai-2015-multi-atlas-abdomen-labeling"},"category":"Computer Vision","categories":["Computer Vision","Medical"],"category_note":null,"description":"**Medical Image Segmentation** is a computer vision task that involves dividing an medical image into multiple segments, where each segment represents a different object or structure of interest in the image. The goal of medical image segmentation is to provide a precise and accurate representation of the objects of interest within the image, typically for the purpose of diagnosis, treatment planning, and quantitative analysis.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [IVD-Net](https://github.com/josedolz/IVD-Net) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Avg DSC","Avg HD"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Avg DSC":null,"Avg HD":null}},"counts":{"rows":23,"rows_with_code":23,"rows_with_paper_page":23,"rows_dated":23,"rows_using_additional_data":9},"rows":[{"rank_in_archive_order":1,"model":"Interactive AI-SAM gt box","metrics":{"Avg DSC":"90.66"},"uses_additional_data":true,"paper_date":"2023-12-05","paper":"/paper/ai-sam-automatic-and-interactive-segment","paper_url":"https://arxiv.org/abs/2312.03119v1","paper_title":"AI-SAM: Automatic and Interactive Segment Anything Model","code":"https://github.com/ymp5078/ai-sam","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"Medical SAM Adapter","metrics":{"Avg DSC":"89.80"},"uses_additional_data":false,"paper_date":"2023-04-25","paper":"/paper/medical-sam-adapter-adapting-segment-anything","paper_url":"https://arxiv.org/abs/2304.12620v7","paper_title":"Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation","code":"https://github.com/supermedintel/medical-sam-adapter","n_code_links":5,"syntology":null},{"rank_in_archive_order":3,"model":"MedSegDiff-v2","metrics":{"Avg DSC":"89.50"},"uses_additional_data":false,"paper_date":"2023-01-19","paper":"/paper/medsegdiff-v2-diffusion-based-medical-image","paper_url":"https://arxiv.org/abs/2301.11798v2","paper_title":"MedSegDiff-V2: Diffusion based Medical Image Segmentation with Transformer","code":"https://github.com/kidswithtokens/medsegdiff","n_code_links":2,"syntology":{"n_ran":13,"n_unverified":8,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"nnUNet","metrics":{"Avg DSC":"88.80","Avg HD":"10.78"},"uses_additional_data":false,"paper_date":"2018-09-27","paper":"/paper/nnu-net-self-adapting-framework-for-u-net","paper_url":"http://arxiv.org/abs/1809.10486v1","paper_title":"nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation","code":"https://github.com/NVIDIA/DeepLearningExamples/tree/ddbcd54056e8d1bc1c4d5a8ab34cb570ebea1947/PyTorch/Segmentation/nnUNet","n_code_links":9,"syntology":{"n_ran":3,"n_unverified":8,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"MedNeXt-L (5x5x5)","metrics":{"Avg DSC":"88.76"},"uses_additional_data":false,"paper_date":"2023-03-17","paper":"/paper/mednext-transformer-driven-scaling-of","paper_url":"https://arxiv.org/abs/2303.09975v5","paper_title":"MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation","code":"https://github.com/MIC-DKFZ/MedNeXt","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":2,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"MIST","metrics":{"Avg DSC":"86.92","Avg HD":"11.07"},"uses_additional_data":false,"paper_date":"2023-10-30","paper":"/paper/mist-medical-image-segmentation-transformer","paper_url":"https://arxiv.org/abs/2310.19898v1","paper_title":"MIST: Medical Image Segmentation Transformer with Convolutional Attention Mixing (CAM) Decoder","code":"https://github.com/rahman-motiur/mist","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"nnFormer","metrics":{"Avg DSC":"86.57","Avg HD":"10.63"},"uses_additional_data":true,"paper_date":"2021-09-07","paper":"/paper/nnformer-interleaved-transformer-for","paper_url":"https://arxiv.org/abs/2109.03201v6","paper_title":"nnFormer: Interleaved Transformer for Volumetric Segmentation","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"AgileFormer","metrics":{"Avg DSC":"86.11","Avg HD":"12.88"},"uses_additional_data":false,"paper_date":"2024-03-29","paper":"/paper/agileformer-spatially-agile-transformer-unet","paper_url":"https://arxiv.org/abs/2404.00122v2","paper_title":"AgileFormer: Spatially Agile Transformer UNet for Medical Image Segmentation","code":"https://github.com/sotiraslab/AgileFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"MERIT","metrics":{"Avg DSC":"84.90","Avg HD":"13.22"},"uses_additional_data":false,"paper_date":"2023-03-29","paper":"/paper/multi-scale-hierarchical-vision-transformer-1","paper_url":"https://arxiv.org/abs/2303.16892v1","paper_title":"Multi-scale Hierarchical Vision Transformer with Cascaded Attention Decoding for Medical Image Segmentation","code":"https://github.com/SLDGroup/MERIT","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"Automatic AI-SAM","metrics":{"Avg DSC":"84.21"},"uses_additional_data":true,"paper_date":"2023-12-05","paper":"/paper/ai-sam-automatic-and-interactive-segment","paper_url":"https://arxiv.org/abs/2312.03119v1","paper_title":"AI-SAM: Automatic and Interactive Segment Anything Model","code":"https://github.com/ymp5078/ai-sam","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"ParaTransCNN","metrics":{"Avg DSC":"83.86","Avg HD":"15.86"},"uses_additional_data":true,"paper_date":"2024-01-27","paper":"/paper/paratranscnn-parallelized-transcnn-encoder","paper_url":"https://arxiv.org/abs/2401.15307v1","paper_title":"ParaTransCNN: Parallelized TransCNN Encoder for Medical Image Segmentation","code":"https://github.com/hongkunsun/paratranscnn","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"EMCAD","metrics":{"Avg DSC":"83.63","Avg HD":"15.68"},"uses_additional_data":false,"paper_date":"2024-05-11","paper":"/paper/emcad-efficient-multi-scale-convolutional","paper_url":"https://arxiv.org/abs/2405.06880v1","paper_title":"EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation","code":"https://github.com/sldgroup/emcad","n_code_links":1,"syntology":{"n_ran":20,"n_unverified":3,"n_samples":23,"n_pointer_only_licence":23}},{"rank_in_archive_order":13,"model":"PAG-TransYnet","metrics":{"Avg DSC":"83.43","Avg HD":"15.82"},"uses_additional_data":false,"paper_date":"2024-04-28","paper":"/paper/rethinking-attention-gated-with-hybrid-dual","paper_url":"https://arxiv.org/abs/2404.18199v1","paper_title":"Rethinking Attention Gated with Hybrid Dual Pyramid Transformer-CNN for Generalized Segmentation in Medical Imaging","code":"https://github.com/faresbougourzi/pagtransynet","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"SegFormer3D","metrics":{"Avg DSC":"82.15"},"uses_additional_data":false,"paper_date":"2024-04-15","paper":"/paper/segformer3d-an-efficient-transformer-for-3d","paper_url":"https://arxiv.org/abs/2404.10156v2","paper_title":"SegFormer3D: an Efficient Transformer for 3D Medical Image Segmentation","code":"https://github.com/osupcvlab/segformer3d","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":15,"model":"MISSFormer","metrics":{"Avg DSC":"81.96","Avg HD":"18.20"},"uses_additional_data":true,"paper_date":"2021-09-15","paper":"/paper/missformer-an-effective-medical-image","paper_url":"https://arxiv.org/abs/2109.07162v1","paper_title":"MISSFormer: An Effective Medical Image Segmentation Transformer","code":"https://github.com/zhifangdeng/missformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"TransUNet","metrics":{"Avg DSC":"81.19"},"uses_additional_data":false,"paper_date":"2024-12-17","paper":"/paper/s2s2-semantic-stacking-for-robust-semantic","paper_url":"https://arxiv.org/abs/2412.13156v1","paper_title":"S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging","code":"https://github.com/ymp5078/semantic-stacking","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"SelfReg-UNet: SwinUNet","metrics":{"Avg DSC":"80.54"},"uses_additional_data":false,"paper_date":"2024-06-21","paper":"/paper/selfreg-unet-self-regularized-unet-for","paper_url":"https://arxiv.org/abs/2406.14896v1","paper_title":"SelfReg-UNet: Self-Regularized UNet for Medical Image Segmentation","code":"https://github.com/chongqingnosubway/selfreg-unet","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"SelfReg-UNet: Vanilla UNet","metrics":{"Avg DSC":"80.34"},"uses_additional_data":false,"paper_date":"2024-06-21","paper":"/paper/selfreg-unet-self-regularized-unet-for","paper_url":"https://arxiv.org/abs/2406.14896v1","paper_title":"SelfReg-UNet: Self-Regularized UNet for Medical Image Segmentation","code":"https://github.com/chongqingnosubway/selfreg-unet","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"FCB Former","metrics":{"Avg DSC":"80.26"},"uses_additional_data":false,"paper_date":"2022-07-16","paper":"/paper/adaptive-t-vmf-dice-loss-for-multi-class","paper_url":"https://arxiv.org/abs/2207.07842v1","paper_title":"Adaptive t-vMF Dice Loss for Multi-class Medical Image Segmentation","code":"https://github.com/usagisukisuki/adaptive_t-vmf_dice_loss","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"SETR","metrics":{"Avg DSC":"79.60"},"uses_additional_data":true,"paper_date":"2020-12-31","paper":"/paper/rethinking-semantic-segmentation-from-a","paper_url":"https://arxiv.org/abs/2012.15840v3","paper_title":"Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":5,"syntology":null},{"rank_in_archive_order":21,"model":"SwinUnet","metrics":{"Avg DSC":"79.13","Avg HD":"21.55"},"uses_additional_data":true,"paper_date":"2021-05-12","paper":"/paper/swin-unet-unet-like-pure-transformer-for","paper_url":"https://arxiv.org/abs/2105.05537v1","paper_title":"Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation","code":"https://github.com/HuCaoFighting/Swin-Unet","n_code_links":7,"syntology":{"n_ran":3,"n_unverified":15,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"UCTransNet","metrics":{"Avg DSC":"78.99","Avg HD":"30.29"},"uses_additional_data":true,"paper_date":"2021-09-09","paper":"/paper/uctransnet-rethinking-the-skip-connections-in","paper_url":"https://arxiv.org/abs/2109.04335v3","paper_title":"UCTransNet: Rethinking the Skip Connections in U-Net from a Channel-wise Perspective with Transformer","code":"https://github.com/lucidrains/imagen-pytorch","n_code_links":3,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":23,"model":"TransUNet","metrics":{"Avg DSC":"77.48","Avg HD":"31.69"},"uses_additional_data":true,"paper_date":"2021-02-08","paper":"/paper/transunet-transformers-make-strong-encoders","paper_url":"https://arxiv.org/abs/2102.04306v1","paper_title":"TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation","code":"https://github.com/PaddlePaddle/PaddleSeg","n_code_links":22,"syntology":{"n_ran":6,"n_unverified":1,"n_samples":7,"n_pointer_only_licence":7}}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":9,"rows_with_any_sample_ran":9,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":9,"samples_over_distinct_papers":{"n_ran":51,"n_unverified":37,"n_samples":88,"n_pointer_only_licence":33,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":51,"n_unverified":37,"n_samples":88,"n_pointer_only_licence":33,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}