{"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/removing-confounding-factors-associated","title":"Removing Confounding Factors Associated Weights in Deep Neural Networks Improves the Prediction Accuracy for Healthcare Applications","arxiv_id":"1803.07276","date":"2018-03-20","proceeding":null,"authors":["Haohan Wang","Zhenglin Wu","Eric P. Xing"],"abstract":"The proliferation of healthcare data has brought the opportunities of\napplying data-driven approaches, such as machine learning methods, to assist\ndiagnosis. Recently, many deep learning methods have been shown with impressive\nsuccesses in predicting disease status with raw input data. However, the\n\"black-box\" nature of deep learning and the high-reliability requirement of\nbiomedical applications have created new challenges regarding the existence of\nconfounding factors. In this paper, with a brief argument that inappropriate\nhandling of confounding factors will lead to models' sub-optimal performance in\nreal-world applications, we present an efficient method that can remove the\ninfluences of confounding factors such as age or gender to improve the\nacross-cohort prediction accuracy of neural networks. One distinct advantage of\nour method is that it only requires minimal changes of the baseline model's\narchitecture so that it can be plugged into most of the existing neural\nnetworks. We conduct experiments across CT-scan, MRA, and EEG brain wave with\nconvolutional neural networks and LSTM to verify the efficiency of our method.","url_abs":"http://arxiv.org/abs/1803.07276v3","url_pdf":"http://arxiv.org/pdf/1803.07276v3.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":"removing-confounding-factors-associated","repo_url":"https://github.com/HaohanWang/CF","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"eeg-1","task_name":"EEG"},{"task_slug":"eeg","task_name":"Electroencephalogram (EEG)"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.07276","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}