{"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/a-unified-mammogram-analysis-method-via","title":"A Unified Mammogram Analysis Method via Hybrid Deep Supervision","arxiv_id":"1808.10646","date":"2018-08-31","proceeding":null,"authors":["Rongzhao Zhang","Han Zhang","Albert C. S. Chung"],"abstract":"Automatic mammogram classification and mass segmentation play a critical role\nin a computer-aided mammogram screening system. In this work, we present a\nunified mammogram analysis framework for both whole-mammogram classification\nand segmentation. Our model is designed based on a deep U-Net with residual\nconnections, and equipped with the novel hybrid deep supervision (HDS) scheme\nfor end-to-end multi-task learning. As an extension of deep supervision (DS),\nHDS not only can force the model to learn more discriminative features like DS,\nbut also seamlessly integrates segmentation and classification tasks into one\nmodel, thus the model can benefit from both pixel-wise and image-wise\nsupervisions. We extensively validate the proposed method on the widely-used\nINbreast dataset. Ablation study corroborates that pixel-wise and image-wise\nsupervisions are mutually beneficial, evidencing the efficacy of HDS. The\nresults of 5-fold cross validation indicate that our unified model matches\nstate-of-the-art performance on both mammogram segmentation and classification\ntasks, which achieves an average segmentation Dice similarity coefficient (DSC)\nof 0.85 and a classification accuracy of 0.89. The code is available at\nhttps://github.com/angrypudding/hybrid-ds.","url_abs":"http://arxiv.org/abs/1808.10646v1","url_pdf":"http://arxiv.org/pdf/1808.10646v1.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":"a-unified-mammogram-analysis-method-via","repo_url":"https://github.com/angrypudding/hybrid-ds","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"whole-mammogram-classification","task_name":"Whole Mammogram Classification"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}