[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86377-en":3,"doc-seo-86377-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86377,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Accelerated MR Elastography Using Learned Neural Network Representation","A deep-learning reconstruction method accelerates magnetic resonance elastography (MRE) by producing fast, high-resolution results from highly undersampled k-space without relying on high-quality supervised training datasets. The approach models deep neural network representations as a nonlinear extension of linear subspace frameworks, learned via a self-supervised multi-level k-space consistency loss. Elastography-specific magnitude/phase priors enforce anatomical similarity and wave-induced displacement smoothness, validated on 3D gradient-echo spiral and multi-slice spin-echo spiral data.","Accelerated MR Elastography Using Learned Neural Network Representation  \nXi Peng  \nAffiliations:  \nDepartments of Radiology, Biomedical Engineering, Electronic and Computer Engineering, and Mechanical Engineering, University of Iowa, Iowa City, IA 52246  \nRunning title: MRE Image Reconstruction with Learned Network Representation  \nCorrespondence:  \nXi Peng, Ph.D.  \nDepartment of Radiology  \nUniversity of Iowa, Iowa City, IA, 52246 [Email: xi-peng@uiowa.edu](Email: xi-peng@uiowa.edu); [stevep1120@gmail.com](stevep1120@gmail.com) ;  \nManuscript word count: 4278  \nAbstract word count: 185  \nABSTRACT  \nPurpose: To develop a deep-learning method for achieving fast high-resolution MRelastography from highly undersampled data without the need of high-quality training dataset.  \nMethods: We first framed the deep neural network representation as a nonlinear extension of the linear subspace model, then used it to represent and reconstruct MRE image repetitions from undersampled k-space data. The network weights were learned using a multi-level kspace consistent loss in a self-supervised manner. To further enhance reconstruction quality, phase-contrast specific magnitude and phase priors were incorporated, including the similarity of anatomical structures and smoothness of wave-induced harmonic displacement. Experiments were conducted using both 3D gradient-echo spiral and multi-slice spin-echo spiral MRE datasets.  \nResults: Compared to the conventional linear subspace-based approaches, the nonlinear network representation method was able to produce superior image reconstruction with suppressed noise and artifacts from a single in-plane spiral arm per MRE repetition (e.g., total R=10), yielding comparable stiffness estimation to the fully sampled data at 2mm resolution.  \nConclusion: This work demonstrated the feasibility of using deep network representations to model and reconstruct MRE images from highly-undersampled data, a nonlinear extension of the subspace-based approach.  \nKeywords: Deep neural network representation ; linear subspace; MR Elastography;  \n1. INTRODUCTION  \nBrain magnetic resonance elastography (MRE) is a powerful tool to noninvasively access tissue biomechanical properties in vivo, with broad potential in various neuroimaging applications including normal pressure hydrocephalus 1, brain tumor2, neurodegenerative disease3, cognitive function4 and brain development5. Higher spatial resolution is particularly important for identifying heterogenous substances and accurately quantifying stiffer tissues. Achieving highresolution MRE relies on two key factors: fast imaging and robust stiffness inversion. This paper focuses on advancing the fast-imaging component using deep learning-based reconstruction.  \nMRE is a phase-contrast MRI technique and typically requires the acquisition of multiple imaging repetitions for encoding wave propagations in the 3D vector space (e.g., 6 directions) at varying vibration phase-offsets. As a result, achieving high spatial resolution within a practical scan time is challenging. To address this issue, fast scan such as echo-planer-imaging (EPI) and spiral6, along with more efficient motion-encoding strategies have been developed, including SLIM7 (sample interval modulation), Ristretto8, DENSE9, 10 (multiphase Displacement ENcoded Stimulated Echo), Magnetization-Prepared11, and distributed encoding MRE 12. On the other hand, constrained reconstruction from undersampled data has been extensively exploited for fast imaging, with compressed sensing 13, 14 and low-rank modeling 15, 16 being among the most widely adopted approaches leveraging the spatial and temporal redundancies. Additionally, anatomical and phase constraints are among the “early” constrained methods used in various MR applications. For instance, the anatomical similarity of a series of MR images has been utilized for denoising and joint image reconstruction via spatially-adaptive regularization17, 18, generalized series mod","cbCaig3wpjHvRdOa","https://ap.wps.com/l/cbCaig3wpjHvRdOa","pdf",4805430,6,1,27,"English","en",105,"# INTRODUCTION\n# METHODS\n## From Low-rank to Deep Neural Network Representation","[{\"question\":\"What problem does the paper address in MR elastography reconstruction?\",\"answer\":\"It targets the challenge of achieving fast, high-resolution MRE when only highly undersampled k-space data are available and when high-quality supervised training datasets are difficult to obtain.\"},{\"question\":\"How does the proposed method learn the reconstruction network without labeled training data?\",\"answer\":\"It uses a self-supervised multi-level k-space consistency loss to learn network weights, enforcing consistency with acquired k-space measurements.\"},{\"question\":\"What priors are used to improve reconstruction quality?\",\"answer\":\"The method incorporates phase-contrast specific magnitude and phase priors, including similarity of anatomical structures and smoothness of wave-induced harmonic displacement.\"}]",1784211275,68,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"accelerated-mr-elastography-using-learned-neural-network-representation","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/accelerated-mr-elastography-using-learned-neural-network-representation/86377/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the paper address in MR elastography reconstruction?","Question",{"text":76,"@type":77},"It targets the challenge of achieving fast, high-resolution MRE when only highly undersampled k-space data are available and when high-quality supervised training datasets are difficult to obtain.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed method learn the reconstruction network without labeled training data?",{"text":81,"@type":77},"It uses a self-supervised multi-level k-space consistency loss to learn network weights, enforcing consistency with acquired k-space measurements.",{"name":83,"@type":74,"acceptedAnswer":84},"What priors are used to improve reconstruction quality?",{"text":85,"@type":77},"The method incorporates phase-contrast specific magnitude and phase priors, including similarity of anatomical structures and smoothness of wave-induced harmonic displacement.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]