{"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/visualizing-convolutional-networks-for-mri","title":"Visualizing Convolutional Networks for MRI-based Diagnosis of Alzheimer's Disease","arxiv_id":"1808.02874","date":"2018-08-08","proceeding":null,"authors":["Johannes Rieke","Fabian Eitel","Martin Weygandt","John-Dylan Haynes","Kerstin Ritter"],"abstract":"Visualizing and interpreting convolutional neural networks (CNNs) is an\nimportant task to increase trust in automatic medical decision making systems.\nIn this study, we train a 3D CNN to detect Alzheimer's disease based on\nstructural MRI scans of the brain. Then, we apply four different gradient-based\nand occlusion-based visualization methods that explain the network's\nclassification decisions by highlighting relevant areas in the input image. We\ncompare the methods qualitatively and quantitatively. We find that all four\nmethods focus on brain regions known to be involved in Alzheimer's disease,\nsuch as inferior and middle temporal gyrus. While the occlusion-based methods\nfocus more on specific regions, the gradient-based methods pick up distributed\nrelevance patterns. Additionally, we find that the distribution of relevance\nvaries across patients, with some having a stronger focus on the temporal lobe,\nwhereas for others more cortical areas are relevant. In summary, we show that\napplying different visualization methods is important to understand the\ndecisions of a CNN, a step that is crucial to increase clinical impact and\ntrust in computer-based decision support systems.","url_abs":"http://arxiv.org/abs/1808.02874v1","url_pdf":"http://arxiv.org/pdf/1808.02874v1.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":"visualizing-convolutional-networks-for-mri","repo_url":"https://github.com/jrieke/cnn-interpretability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}