{"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/practical-window-setting-optimization-for","title":"Practical Window Setting Optimization for Medical Image Deep Learning","arxiv_id":"1812.00572","date":"2018-12-03","proceeding":null,"authors":["Hyunkwang Lee","Myeongchan Kim","Synho Do"],"abstract":"The recent advancements in deep learning have allowed for numerous\napplications in computed tomography (CT), with potential to improve diagnostic\naccuracy, speed of interpretation, and clinical efficiency. However, the deep\nlearning community has to date neglected window display settings - a key\nfeature of clinical CT interpretation and opportunity for additional\noptimization. Here we propose a window setting optimization (WSO) module that\nis fully trainable with convolutional neural networks (CNNs) to find optimal\nwindow settings for clinical performance. Our approach was inspired by the\nmethod commonly used by practicing radiologists to interpret CT images by\nadjusting window settings to increase the visualization of certain pathologies.\nOur approach provides optimal window ranges to enhance the conspicuity of\nabnormalities, and was used to enable performance enhancement for intracranial\nhemorrhage and urinary stone detection. On each task, the WSO model\noutperformed models trained over the full range of Hounsfield unit values in CT\nimages, as well as images windowed with pre-defined settings. The WSO module\ncan be readily applied to any analysis of CT images, and can be further\ngeneralized to tasks on other medical imaging modalities.","url_abs":"http://arxiv.org/abs/1812.00572v1","url_pdf":"http://arxiv.org/pdf/1812.00572v1.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":"practical-window-setting-optimization-for","repo_url":"https://github.com/Synho/windows_optimization","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"practical-window-setting-optimization-for","repo_url":"https://github.com/MGH-LMIC/windows_optimization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"practical-window-setting-optimization-for","repo_url":"https://github.com/suryachintu/RSNA-Intracranial-Hemorrhage-Detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"computed-tomography-ct","task_name":"Computed Tomography (CT)"},{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"diagnostic","task_name":"Diagnostic"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}