{"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/pwls-ultra-an-efficient-clustering-and","title":"PWLS-ULTRA: An Efficient Clustering and Learning-Based Approach for Low-Dose 3D CT Image Reconstruction","arxiv_id":"1703.09165","date":"2017-03-27","proceeding":null,"authors":["Xuehang Zheng","Saiprasad Ravishankar","Yong Long","Jeffrey A. Fessler"],"abstract":"The development of computed tomography (CT) image reconstruction methods that\nsignificantly reduce patient radiation exposure while maintaining high image\nquality is an important area of research in low-dose CT (LDCT) imaging. We\npropose a new penalized weighted least squares (PWLS) reconstruction method\nthat exploits regularization based on an efficient Union of Learned TRAnsforms\n(PWLS-ULTRA). The union of square transforms is pre-learned from numerous image\npatches extracted from a dataset of CT images or volumes. The proposed\nPWLS-based cost function is optimized by alternating between a CT image\nreconstruction step, and a sparse coding and clustering step. The CT image\nreconstruction step is accelerated by a relaxed linearized augmented Lagrangian\nmethod with ordered-subsets that reduces the number of forward and back\nprojections. Simulations with 2-D and 3-D axial CT scans of the extended\ncardiac-torso phantom and 3D helical chest and abdomen scans show that for both\nnormal-dose and low-dose levels, the proposed method significantly improves the\nquality of reconstructed images compared to PWLS reconstruction with a\nnonadaptive edge-preserving regularizer (PWLS-EP). PWLS with regularization\nbased on a union of learned transforms leads to better image reconstructions\nthan using a single learned square transform. We also incorporate patch-based\nweights in PWLS-ULTRA that enhance image quality and help improve image\nresolution uniformity. The proposed approach achieves comparable or better\nimage quality compared to learned overcomplete synthesis dictionaries, but\nimportantly, is much faster (computationally more efficient).","url_abs":"http://arxiv.org/abs/1703.09165v3","url_pdf":"http://arxiv.org/pdf/1703.09165v3.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":"pwls-ultra-an-efficient-clustering-and","repo_url":"https://github.com/xuehangzheng/PWLS-ULTRA-for-Low-Dose-3D-CT-Image-Reconstruction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"computed-tomography-ct","task_name":"Computed Tomography (CT)"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"}],"methods":[],"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}