{"id":780853,"date":"2021-10-14T10:59:05","date_gmt":"2021-10-14T17:59:05","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-project&#038;p=780853"},"modified":"2021-10-14T10:59:07","modified_gmt":"2021-10-14T17:59:07","slug":"machine-learning-for-image-reconstruction","status":"publish","type":"msr-project","link":"https:\/\/www.microsoft.com\/en-us\/research\/project\/machine-learning-for-image-reconstruction\/","title":{"rendered":"Machine Learning for Image Reconstruction"},"content":{"rendered":"<section class=\"mb-3 moray-highlight\">\n\t<div class=\"card-img-overlay mx-lg-0\">\n\t\t<div class=\"card-background  has-background- card-background--full-bleed\">\n\t\t\t<img loading=\"lazy\" decoding=\"async\" width=\"1400\" height=\"788\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788.jpg\" class=\"attachment-full size-full\" alt=\"Machine Learning Image Reconstruction: image of a brain scan\" style=\"object-position: 59% 50%\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788.jpg 1400w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-300x169.jpg 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-1024x576.jpg 1024w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-768x432.jpg 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-1066x600.jpg 1066w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-655x368.jpg 655w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-343x193.jpg 343w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-240x135.jpg 240w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-640x360.jpg 640w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-960x540.jpg 960w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/BioMed_MLImageReconstruction_10-2021_1400x788-1280x720.jpg 1280w\" sizes=\"auto, (max-width: 1400px) 100vw, 1400px\" \/>\t\t<\/div>\n\t\t<!-- Foreground -->\n\t\t<div class=\"card-foreground d-flex mt-md-n5 my-lg-5 px-g px-lg-0\">\n\t\t\t<!-- Container -->\n\t\t\t<div class=\"container d-flex mt-md-n5 my-lg-5 align-self-center\">\n\t\t\t\t<!-- Card wrapper -->\n\t\t\t\t<div class=\"w-100 w-lg-col-5\">\n\t\t\t\t\t<!-- Card -->\n\t\t\t\t\t<div class=\"card material-md-card py-5 px-md-5\">\n\t\t\t\t\t\t<div class=\"card-body \">\n\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\n\n<h1 id=\"machine-learning-for-image-reconstruction\">Machine Learning for Image Reconstruction<\/h1>\n\n\n\n<p>Making sense of sparse data<\/p>\n\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n<\/section>\n\n\n\n\n\n<div class=\"wp-block-image is-style-default\"><figure class=\"alignright size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"730\" height=\"443\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/Machine-Learning-Image-Reconstruction_graphic.png\" alt=\"machine learning image reconstruction workflow illustration\" class=\"wp-image-782680\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/Machine-Learning-Image-Reconstruction_graphic.png 730w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/Machine-Learning-Image-Reconstruction_graphic-300x182.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2021\/10\/Machine-Learning-Image-Reconstruction_graphic-240x146.png 240w\" sizes=\"auto, (max-width: 730px) 100vw, 730px\" \/><\/figure><\/div>\n\n\n\n<p>The acquisition of raw data for medical images is often time consuming or harmful to patients. Algorithms and techniques that allow us to reconstruct images from datasets that are under-sampled or corrupted by artifacts can enable an improved patient experience and reduce costs. At Microsoft we are engaged in the development of machine learning models that facilitate accelerated imaging. We have enhanced the <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/microsoft\/InnerEye-DeepLearning\" target=\"_blank\" rel=\"noopener noreferrer\">InnerEye<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> toolkit to support training of reconstruction models. Specifically, we are providing dedicated tooling to enable the use of the <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/microsoft\/fastmri-plus\/\" target=\"_blank\" rel=\"noopener noreferrer\">FastMRI<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> dataset and existing model architectures. This includes tooling for managing the large MRI raw datasets needed for training the models. We have also published <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/github.com\/microsoft\/fastmri-plus\" target=\"_blank\" rel=\"noopener noreferrer\">clinical annotations<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> that enable evaluation of new models in the context of pathology.<\/p>\n\n\n","protected":false},"excerpt":{"rendered":"<p>Machine learning-based image reconstruction can be used to reconstruct images from sparse or corrupted datasets. We are developing the necessary techniques and assets for training and evaluation of image reconstruction models.<\/p>\n","protected":false},"featured_media":785077,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"research-area":[13553],"msr-locale":[268875],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-780853","msr-project","type-msr-project","status-publish","has-post-thumbnail","hentry","msr-research-area-medical-health-genomics","msr-locale-en_us","msr-archive-status-active"],"msr_project_start":"","related-publications":[773551],"related-downloads":[],"related-videos":[],"related-groups":[780706],"related-events":[],"related-opportunities":[],"related-posts":[],"related-articles":[],"tab-content":[],"slides":[],"related-researchers":[{"type":"user_nicename","display_name":"Michael Hansen","user_id":40723,"people_section":"Section name 0","alias":"mihansen"},{"type":"user_nicename","display_name":"Anton Schwaighofer","user_id":31059,"people_section":"Section name 0","alias":"antonsc"},{"type":"user_nicename","display_name":"John Stairs","user_id":40858,"people_section":"Section name 0","alias":"jostairs"}],"msr_research_lab":[849856],"msr_impact_theme":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/780853","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-project"}],"version-history":[{"count":11,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/780853\/revisions"}],"predecessor-version":[{"id":850003,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/780853\/revisions\/850003"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/785077"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=780853"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=780853"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=780853"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=780853"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=780853"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}