{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-multi-scale-3d-cnn-with-fully","title":"Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation","arxiv_id":"1603.05959","date":"2016-03-18","proceeding":null,"authors":["Konstantinos Kamnitsas","Christian Ledig","Virginia F. J. Newcombe","Joanna P. Simpson","Andrew D. Kane","David K. Menon","Daniel Rueckert","Ben Glocker"],"abstract":"We propose a dual pathway, 11-layers deep, three-dimensional Convolutional\nNeural Network for the challenging task of brain lesion segmentation. The\ndevised architecture is the result of an in-depth analysis of the limitations\nof current networks proposed for similar applications. To overcome the\ncomputational burden of processing 3D medical scans, we have devised an\nefficient and effective dense training scheme which joins the processing of\nadjacent image patches into one pass through the network while automatically\nadapting to the inherent class imbalance present in the data. Further, we\nanalyze the development of deeper, thus more discriminative 3D CNNs. In order\nto incorporate both local and larger contextual information, we employ a dual\npathway architecture that processes the input images at multiple scales\nsimultaneously. For post-processing of the network's soft segmentation, we use\na 3D fully connected Conditional Random Field which effectively removes false\npositives. Our pipeline is extensively evaluated on three challenging tasks of\nlesion segmentation in multi-channel MRI patient data with traumatic brain\ninjuries, brain tumors, and ischemic stroke. We improve on the state-of-the-art\nfor all three applications, with top ranking performance on the public\nbenchmarks BRATS 2015 and ISLES 2015. Our method is computationally efficient,\nwhich allows its adoption in a variety of research and clinical settings. The\nsource code of our implementation is made publicly available.","url_abs":"http://arxiv.org/abs/1603.05959v3","url_pdf":"http://arxiv.org/pdf/1603.05959v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-multi-scale-3d-cnn-with-fully","repo_url":"https://github.com/etjoa003/medical_imaging","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"efficient-multi-scale-3d-cnn-with-fully","repo_url":"https://github.com/Kamnitsask/deepmedic","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"3d-medical-imaging-segmentation","task_name":"3D Medical Imaging Segmentation"},{"task_slug":"brain-lesion-segmentation-from-mri","task_name":"Brain Lesion Segmentation From Mri"},{"task_slug":"brain-tumor-segmentation","task_name":"Brain Tumor Segmentation"},{"task_slug":"lesion-segmentation","task_name":"Lesion Segmentation"},{"task_slug":"medical-image-segmentation","task_name":"Medical Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/brain-tumor-segmentation-on-brats-2015","task":"Brain Tumor Segmentation","dataset":"BRATS-2015","model":"3D CNN + CRF","rank_in_archive_order":3,"of":4,"metrics":{"Dice Score":"85.0%"},"uses_additional_data":true},{"leaderboard":"/sota/lesion-segmentation-on-isles-2015","task":"Lesion Segmentation","dataset":"ISLES-2015","model":"3D CNN + CRF","rank_in_archive_order":1,"of":1,"metrics":{"Dice Score":"59.0%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.05959","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1603.05959"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/etjoa003/medical_imaging","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Kamnitsask/deepmedic","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"unverified":12},"by_repo_kind":{"listed":{"samples":12,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"b79ab7d1e72f7b3b","entry":"augment_imgs_of_case","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/dataManagement/augmentImage.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/dataManagement/augmentImage.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"b79ab7d1e72f7b3b"}},{"code_sha256_prefix":"8db9843f9dcbf528","entry":"calc_rec_field_of_path_assuming_strides_1","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/utils.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"8db9843f9dcbf528"}},{"code_sha256_prefix":"338063673d94ab09","entry":"check_kern_dims_per_l_correct_3d_and_n_layers","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/utils.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"338063673d94ab09"}},{"code_sha256_prefix":"8aecaa717071051e","entry":"check_rec_field_vs_inp_dims","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/utils.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"8aecaa717071051e"}},{"code_sha256_prefix":"4357cf5f087db495","entry":"conv_3d","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/ops.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"4357cf5f087db495"}},{"code_sha256_prefix":"9d5e70582a1d30c1","entry":"dsc","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/cost_functions.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/cost_functions.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9d5e70582a1d30c1"}},{"code_sha256_prefix":"730ced37040cedf6","entry":"iou","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/cost_functions.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/cost_functions.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"730ced37040cedf6"}},{"code_sha256_prefix":"a6fcd9282f9d659f","entry":"prelu","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/ops.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a6fcd9282f9d659f"}},{"code_sha256_prefix":"346f2897c0851202","entry":"random_affine_deformation","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/dataManagement/augmentImage.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/dataManagement/augmentImage.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"346f2897c0851202"}},{"code_sha256_prefix":"9a505431f15e3e27","entry":"random_gamma_correction","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/dataManagement/augmentImage.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/dataManagement/augmentImage.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9a505431f15e3e27"}},{"code_sha256_prefix":"5d605f64da7a03e1","entry":"relu","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/ops.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"5d605f64da7a03e1"}},{"code_sha256_prefix":"21d9f65263774216","entry":"x_entr","repo":"Kamnitsask/deepmedic","repo_kind":"listed","path":"deepmedic/neuralnet/cost_functions.py","file_url":"https://github.com/Kamnitsask/deepmedic/blob/HEAD/deepmedic/neuralnet/cost_functions.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"21d9f65263774216"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}