{"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/brain-tumor-segmentation-with-deep-neural","title":"Brain Tumor Segmentation with Deep Neural Networks","arxiv_id":"1505.03540","date":"2015-05-13","proceeding":null,"authors":["Mohammad Havaei","Axel Davy","David Warde-Farley","Antoine Biard","Aaron Courville","Yoshua Bengio","Chris Pal","Pierre-Marc Jodoin","Hugo Larochelle"],"abstract":"In this paper, we present a fully automatic brain tumor segmentation method\nbased on Deep Neural Networks (DNNs). The proposed networks are tailored to\nglioblastomas (both low and high grade) pictured in MR images. By their very\nnature, these tumors can appear anywhere in the brain and have almost any kind\nof shape, size, and contrast. These reasons motivate our exploration of a\nmachine learning solution that exploits a flexible, high capacity DNN while\nbeing extremely efficient. Here, we give a description of different model\nchoices that we've found to be necessary for obtaining competitive performance.\nWe explore in particular different architectures based on Convolutional Neural\nNetworks (CNN), i.e. DNNs specifically adapted to image data.\n  We present a novel CNN architecture which differs from those traditionally\nused in computer vision. Our CNN exploits both local features as well as more\nglobal contextual features simultaneously. Also, different from most\ntraditional uses of CNNs, our networks use a final layer that is a\nconvolutional implementation of a fully connected layer which allows a 40 fold\nspeed up. We also describe a 2-phase training procedure that allows us to\ntackle difficulties related to the imbalance of tumor labels. Finally, we\nexplore a cascade architecture in which the output of a basic CNN is treated as\nan additional source of information for a subsequent CNN. Results reported on\nthe 2013 BRATS test dataset reveal that our architecture improves over the\ncurrently published state-of-the-art while being over 30 times faster.","url_abs":"http://arxiv.org/abs/1505.03540v3","url_pdf":"http://arxiv.org/pdf/1505.03540v3.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":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/AchintyaX/Brain_tumor_segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/IAmSuyogJadhav/Brainy","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/RobinRajSB/Brain-Tumor-Segmentation-Using-CNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/abhi134/Brain_Tumor_Segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/ajinas-ibrahim/brain_tumor","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/dijju/mri-cnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/jadevaibhav/Brain-Tumor-Segmentation-using-Deep-Neural-networks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/mnsv73/brain-tumour-segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/naldeborgh7575/brain_segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/peteraugustine/seg3","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/shalabh147/Brain-Tumor-Segmentation-and-Survival-Prediction-using-Deep-Neural-Networks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/AryaKoureshi/Brain-tumor-detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/tbymiracle/Brain-Tumor-Segmentation-Paddle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok"}},{"paper_slug":"brain-tumor-segmentation-with-deep-neural","repo_url":"https://github.com/yangyucheng000/Paper-3/tree/main/msBraVL-master","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"brain-tumor-segmentation","task_name":"Brain Tumor Segmentation"},{"task_slug":"medical-image-segmentation","task_name":"Medical Image Segmentation"},{"task_slug":"tumor-segmentation","task_name":"Tumor Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/brain-tumor-segmentation-on-brats-2013","task":"Brain Tumor Segmentation","dataset":"BRATS-2013","model":"InputCascadeCNN","rank_in_archive_order":3,"of":3,"metrics":{"Dice Score":"0.88"},"uses_additional_data":false},{"leaderboard":"/sota/brain-tumor-segmentation-on-brats-2013-1","task":"Brain Tumor Segmentation","dataset":"BRATS-2013 leaderboard","model":"InputCascadeCNN","rank_in_archive_order":1,"of":2,"metrics":{"Dice Score":"0.84"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1505.03540","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}