[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123777-en":3,"doc-seo-123777-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123777,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",7,"Healthcare","Optimisation of quantitative brain diffusion-relaxation MRI acquisition protocols with physics-informed machine learning","Diffusion-relaxation MRI aims to derive quantitative microstructural tissue descriptors such as orientation, size, and shape, yet clinical use is limited by long acquisition times. The study introduces a physics-informed learning framework that selects an optimal subset of diffusion-relaxation measurements to shorten protocols, predicts non-measured signals, and estimates quantitative parameters. Training and validation use in vivo and synthetic 5D diffusion-relaxation MRI data from five healthy subjects; testing uses a sixth participant, comparing physics-informed and data-driven models against manual selection and Cramér–Rao bound optimisation.","Medical Image Analysis 94 (2024) 103134  \n| Optimisation of quantitative brain diffusion-relaxation MRI acquisition protocols with physics-informed machine learning\u003Cbr>Álvaro Planchuelo-Gómez a,b, Maxime Descoteaux c, Hugo Larochelle d, Jana Hutter e, Derek K. Jones a, Chantal M.W. Tax f,g,∗\u003Cbr>a Cardiff University Brain Research Imaging Centre (CUBRIC), School of Psychology, Cardiff University, Cardiff, United Kingdom b Imaging Processing Laboratory, Universidad de Valladolid, Valladolid, Spain\u003Cbr>c Sherbrooke Connectivity Imaging Lab (SCIL), Computer Science Department, Université de Sherbrooke, Sherbrooke, QC, Canada d Google DeepMind, Montréal, QC, Canada\u003Cbr>e Centre for Medical Engineering, Centre for the Developing Brain, King’s College London, London, United Kingdom f Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands\u003Cbr>g Cardiff University Brain Research Imaging Centre (CUBRIC), School of Physics and Astronomy, Cardiff University, Cardiff, United Kingdom |  |  |\n| --- | --- | --- |\n| A R T I C L E I N F O |  | A B S T R A C T |\n| Keywords: Quantitative MRI Machine learning Brain\u003Cbr>Diffusion-relaxation |  | Diffusion-relaxation MRI aims to extract quantitative measures that characterise microstructural tissue properties such as orientation, size, and shape, but long acquisition times are typically required. This work proposesa physics-informed learning framework to extract an optimal subset of diffusion-relaxation MRI measurements for enabling shorter acquisition times, predict non-measured signals, and estimate quantitative parameters.\u003Cbr>In vivo and synthetic brain 5D-Diffusion-􀁔1-􀁔2∗ -weighted MRI data obtained from five healthy subjects were used for training and validation, and from a sixth participant for testing. One fully data-driven and two physicsinformed machine learning methods were implemented and compared to two manual selection procedures and Cramér–Rao lower bound optimisation.\u003Cbr>The physics-informed approaches could identify measurement-subsets that yielded more consistently accurate parameter estimates in simulations than other approaches, with similar signal prediction error. Fivefold shorter protocols yielded error distributions of estimated quantitative parameters with very small effect sizes compared to estimates from the full protocol. Selected subsets commonly included a denser sampling of the shortest and longest inversion time, lowest echo time, and high b-value.\u003Cbr>The proposed framework combining machine learning and MRI physics offers a promising approach to develop shorter imaging protocols without compromising the quality of parameter estimates and signal predictions. |\n\n1. Introduction  \nClinical magnetic resonance images (MRI) typically show qualitative tissue contrast, with intensities arbitrarily scaled according to different MR-phenomena. Quantitative MRI, on the other hand, aims to extract reproducible measures more directly related to tissue properties that can be studied longitudinally and/or across populations. Examples of the phenomena that can be studied to assess tissue changes quantitatively in health and disease are diffusion and relaxation (Cercignani and Bouyagoub, 2018).  \nTo quantify tissue properties, distinct MRI experiments are typically performed to probe and subsequently quantify each individual phenomenon by acquiring multiple images in which one or a few experimental MRI parameters are varied. Diffusion MRI (dMRI) sensitises  \nthe MRI signal to the random molecular motion of water by acquiring MR data with externally applied field gradients with different amplitudes and along different orientations. Relaxation MRI measures the return to equilibrium of nuclear spin polarisation after a radiofrequency pulse, and relaxation times can provide valuable insights into tissue composition and pathology. Notwithstanding the ultimate promise of quantitative diffusion and relaxation measures to improve diagnosis, disease monitoring, and un","cbCaid71464C8oAc","https://ap.wps.com/l/cbCaid71464C8oAc","pdf",6087362,1,15,"English","en",105,"# Introduction\n## Quantitative MRI and its rationale\n## Diffusion and relaxation as complementary contrasts\n## Limits of long acquisitions and the need for joint optimization\n# Methods and learning framework\n## Physics-informed measurement subset selection\n## Signal prediction and parameter estimation\n## Compared baselines and optimisation references\n# Experimental setup and evaluation\n## Training, validation, and test data\n## Performance in simulations and in vivo results\n## Error distributions and protocol shortening impact\n# Results and discussion\n## Measurement subsets yielding accurate parameters\n## Trade-offs between protocol length, accuracy, and signal error\n# Conclusion\n## Physics-informed MRI physics meets machine learning for shorter protocols","[{\"question\":\"What problem does diffusion-relaxation MRI address, and why is acquisition time a limitation?\",\"answer\":\"It extracts quantitative measures of microstructural tissue properties, but achieving accurate parameter estimates typically requires long acquisition times that hinder clinical adoption.\"},{\"question\":\"How does the proposed framework reduce scan time while preserving parameter quality?\",\"answer\":\"It uses physics-informed machine learning to identify an optimal subset of diffusion-relaxation measurements, predict missing signals, and estimate quantitative parameters from fewer acquisitions.\"},{\"question\":\"What data and comparison methods were used to validate the approach?\",\"answer\":\"Training and validation used in vivo and synthetic 5D diffusion-relaxation MRI data from five healthy subjects, with testing on a sixth participant. The method was compared with one data-driven approach, two physics-informed approaches, two manual selection procedures, and Cramér–Rao lower bound optimisation.\"}]","Optimisation of quantitative brain diffusion-relaxation MRI acquisition protocols with physics-informed machine learning | PDF",1785818511,38,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"optimisation-of-quantitative-brain-diffusion-relaxation-mri-acquisition-protocols-with-physics-informed-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"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/healthcare/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/optimisation-of-quantitative-brain-diffusion-relaxation-mri-acquisition-protocols-with-physics-informed-machine-learning/123777/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does diffusion-relaxation MRI address, and why is acquisition time a limitation?","Question",{"text":75,"@type":76},"It extracts quantitative measures of microstructural tissue properties, but achieving accurate parameter estimates typically requires long acquisition times that hinder clinical adoption.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed framework reduce scan time while preserving parameter quality?",{"text":80,"@type":76},"It uses physics-informed machine learning to identify an optimal subset of diffusion-relaxation measurements, predict missing signals, and estimate quantitative parameters from fewer acquisitions.",{"name":82,"@type":73,"acceptedAnswer":83},"What data and comparison methods were used to validate the approach?",{"text":84,"@type":76},"Training and validation used in vivo and synthetic 5D diffusion-relaxation MRI data from five healthy subjects, with testing on a sixth participant. The method was compared with one data-driven approach, two physics-informed approaches, two manual selection procedures, and Cramér–Rao lower bound optimisation.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,118,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":116,"slug":117},40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",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":106,"slug":138},19,"General","general"]