{"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/a-deep-cascade-of-convolutional-neural","title":"A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction","arxiv_id":"1704.02422","date":"2017-04-08","proceeding":null,"authors":["Jo Schlemper","Jose Caballero","Joseph V. Hajnal","Anthony Price","Daniel Rueckert"],"abstract":"Inspired by recent advances in deep learning, we propose a framework for\nreconstructing dynamic sequences of 2D cardiac magnetic resonance (MR) images\nfrom undersampled data using a deep cascade of convolutional neural networks\n(CNNs) to accelerate the data acquisition process. In particular, we address\nthe case where data is acquired using aggressive Cartesian undersampling.\nFirstly, we show that when each 2D image frame is reconstructed independently,\nthe proposed method outperforms state-of-the-art 2D compressed sensing\napproaches such as dictionary learning-based MR image reconstruction, in terms\nof reconstruction error and reconstruction speed. Secondly, when reconstructing\nthe frames of the sequences jointly, we demonstrate that CNNs can learn\nspatio-temporal correlations efficiently by combining convolution and data\nsharing approaches. We show that the proposed method consistently outperforms\nstate-of-the-art methods and is capable of preserving anatomical structure more\nfaithfully up to 11-fold undersampling. Moreover, reconstruction is very fast:\neach complete dynamic sequence can be reconstructed in less than 10s and, for\nthe 2D case, each image frame can be reconstructed in 23ms, enabling real-time\napplications.","url_abs":"http://arxiv.org/abs/1704.02422v2","url_pdf":"http://arxiv.org/pdf/1704.02422v2.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":"a-deep-cascade-of-convolutional-neural","repo_url":"https://github.com/js3611/Deep-MRI-Reconstruction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-deep-cascade-of-convolutional-neural","repo_url":"https://github.com/cjandrioli/Deep-MRI-Reconstruction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-deep-cascade-of-convolutional-neural","repo_url":"https://github.com/myyaqubpython/https-github.com-cq615-Deep-MRI-Reconstruction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"a-deep-cascade-of-convolutional-neural","repo_url":"https://github.com/sainzmac/Deep-MRI-Reconstruction-master","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"dictionary-learning","task_name":"Dictionary Learning"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"},{"task_slug":"compressed-sensing","task_name":"compressed sensing"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.02422","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}