{"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/can-deep-learning-outperform-modern","title":"Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?","arxiv_id":"1811.03691","date":"2018-11-08","proceeding":null,"authors":["Hongming Shan","Atul Padole","Fatemeh Homayounieh","Uwe Kruger","Ruhani Doda Khera","Chayanin Nitiwarangkul","Mannudeep K. Kalra","Ge Wang"],"abstract":"Commercial iterative reconstruction techniques on modern CT scanners target\nradiation dose reduction but there are lingering concerns over their impact on\nimage appearance and low contrast detectability. Recently, machine learning,\nespecially deep learning, has been actively investigated for CT. Here we design\na novel neural network architecture for low-dose CT (LDCT) and compare it with\ncommercial iterative reconstruction methods used for standard of care CT. While\npopular neural networks are trained for end-to-end mapping, driven by big data,\nour novel neural network is intended for end-to-process mapping so that\nintermediate image targets are obtained with the associated search gradients\nalong which the final image targets are gradually reached. This learned dynamic\nprocess allows to include radiologists in the training loop to optimize the\nLDCT denoising workflow in a task-specific fashion with the denoising depth as\na key parameter. Our progressive denoising network was trained with the Mayo\nLDCT Challenge Dataset, and tested on images of the chest and abdominal regions\nscanned on the CT scanners made by three leading CT vendors. The best deep\nlearning based reconstructions are systematically compared to the best\niterative reconstructions in a double-blinded reader study. It is found that\nour deep learning approach performs either comparably or favorably in terms of\nnoise suppression and structural fidelity, and runs orders of magnitude faster\nthan the commercial iterative CT reconstruction algorithms.","url_abs":"http://arxiv.org/abs/1811.03691v1","url_pdf":"http://arxiv.org/pdf/1811.03691v1.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":"can-deep-learning-outperform-modern","repo_url":"https://github.com/hmshan/MAP-NN","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"ct-reconstruction","task_name":"CT Reconstruction"},{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.03691","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1811.03691"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hmshan/MAP-NN","reach":null}],"summary":{"ran_fixture":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"44654289537e4ae8","entry":"normalization","repo":"hmshan/MAP-NN","repo_kind":"official","path":"MAP_NN_training.py","file_url":"https://github.com/hmshan/MAP-NN/blob/HEAD/MAP_NN_training.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"44654289537e4ae8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}