{"id":{"repo_id":"cornell","oai_identifier":"oai:ecommons.cornell.edu:1813/67384"},"canonical_url":"https://search.ndltd.org/etd/cornell/oai:ecommons.cornell.edu:1813/67384","repository":{"repo_id":"cornell","name":"Cornell University","base_url":"https://ecommons.cornell.edu/server/oai/request"},"display":{"title":"Learning-Based Optimization of the Under-Sampling Pattern in MRI","abstract":"The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE .","abstract_html":"The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE .","abstract_has_math":false,"creators":["Bahadir, Cagla Deniz"],"institution":"Cornell University","degree_name":"M.S., Biomedical Engineering","degree_level":"Master of Science","degree_discipline":"Biomedical Engineering","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":["Sabuncu, Mert"],"committee_members":["Doerschuk, Peter","Wang, Yi"],"year":2019,"date_issued":"2019-05-30","date_published":"2019-05-30","updated_at":"2026-11-19T03:20:35Z","subjects":["Computer engineering","Electrical engineering","neural networks","Optimization","Magnetic Resonance Imaging","Biomedical engineering","Compressed Sensing","Under-sampling","machine learning"],"languages":["en_US"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/q2y0-rn35"],"render_values":[{"text":"https://doi.org/10.7298/q2y0-rn35","href":"https://doi.org/10.7298/q2y0-rn35","code":true}]},{"key":"dc:identifier.bibid","label":"Dc Identifier Bibid","values":["11050366"],"render_values":[{"text":"11050366","href":null,"code":true}]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 10489","ProQuest Publication ID: 13864844"],"render_values":[{"text":"ProQuest Submission ID: 10489","href":null,"code":true},{"text":"ProQuest Publication ID: 13864844","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1813/67384","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/1813/67384","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://ecommons.cornell.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aecommons.cornell.edu%3A1813%2F67384","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.chair","label":"Chair","values":["Sabuncu, Mert"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Doerschuk, Peter","Wang, Yi"]},{"key":"dc:creator","label":"Author","values":["Bahadir, Cagla Deniz"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2019-10-15T15:30:46Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2019-12-05T07:00:50Z"]},{"key":"dc:date.issued","label":"Date","values":["2019-05-30"]},{"key":"dc:type","label":"Dc Type","values":["dissertation or thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Biomedical Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Master of Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S., Biomedical Engineering"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Cornell University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computer engineering","Electrical engineering","neural networks","Optimization","Magnetic Resonance Imaging","Biomedical engineering","Compressed Sensing","Under-sampling","machine learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.bibid","label":"Dc Identifier Bibid","values":["11050366"]},{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/q2y0-rn35"]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 10489","ProQuest Publication ID: 13864844"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1813/67384"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"cerif:openaire","label":"Cerif Openaire","values":["<Publication xmlns=\"https://www.openaire.eu/cerif-profile/1.1/\" id=\"30021b49-7106-40c3-bac7-2af77dbc3816\"> <Type xmlns=\"https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types\">http://purl.org/coar/resource_type/c_1843</Type> <Language>en_US</Language> <Title>Learning-Based Optimization of the Under-Sampling Pattern in MRI</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019-05-30</PublicationDate> <DOI>https://doi.org/10.7298/q2y0-rn35</DOI> <Authors> <Author> <DisplayName>Bahadir, Cagla Deniz</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <Keyword>Computer engineering</Keyword> <Keyword>Electrical engineering</Keyword> <Keyword>neural networks</Keyword> <Keyword>Optimization</Keyword> <Keyword>Magnetic Resonance Imaging</Keyword> <Keyword>Biomedical engineering</Keyword> <Keyword>Compressed Sensing</Keyword> <Keyword>Under-sampling</Keyword> <Keyword>machine learning</Keyword> <Abstract>The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE .</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"]},{"key":"cris:virtual.author","label":"Cris Virtual Author","values":["Bahadir, Cagla Deniz"]},{"key":"cris:virtualsource.collection","label":"Cris Virtualsource Collection","values":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"]},{"key":"dc:description.abstract","label":"Abstract","values":["The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE ."]},{"key":"dc:description.collection","label":"Dc Description Collection","values":["Cornell Theses and Dissertations"]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dcterms:license","label":"Dcterms License","values":["https://hdl.handle.net/1813/59810"]},{"key":"dc:title","label":"Title","values":["Learning-Based Optimization of the Under-Sampling Pattern in MRI"]},{"key":"dspace:entity.type","label":"Dspace Entity Type","values":["Publication"]},{"key":"others:access_status","label":"Others Access Status","values":["open.access"]}]}],"canonical_facts":{"cerif:openaire":["<Publication xmlns=\"https://www.openaire.eu/cerif-profile/1.1/\" id=\"30021b49-7106-40c3-bac7-2af77dbc3816\"> <Type xmlns=\"https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types\">http://purl.org/coar/resource_type/c_1843</Type> <Language>en_US</Language> <Title>Learning-Based Optimization of the Under-Sampling Pattern in MRI</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019-05-30</PublicationDate> <DOI>https://doi.org/10.7298/q2y0-rn35</DOI> <Authors> <Author> <DisplayName>Bahadir, Cagla Deniz</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <Keyword>Computer engineering</Keyword> <Keyword>Electrical engineering</Keyword> <Keyword>neural networks</Keyword> <Keyword>Optimization</Keyword> <Keyword>Magnetic Resonance Imaging</Keyword> <Keyword>Biomedical engineering</Keyword> <Keyword>Compressed Sensing</Keyword> <Keyword>Under-sampling</Keyword> <Keyword>machine learning</Keyword> <Abstract>The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE .</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"],"cris:virtual.author":["Bahadir, Cagla Deniz"],"cris:virtualsource.collection":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"],"dc:contributor.chair":["Sabuncu, Mert"],"dc:contributor.committeemember":["Doerschuk, Peter","Wang, Yi"],"dc:creator":["Bahadir, Cagla Deniz"],"dc:date.accessioned":["2019-10-15T15:30:46Z"],"dc:date.available":["2019-12-05T07:00:50Z"],"dc:date.issued":["2019-05-30"],"dc:description.abstract":["The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain). In this thesis, we focus on the optimization of the sub-sampling pattern with a data-driven framework. Since the reconstruction quality of the models are shown to be strongly dependent on the sub-sampling pattern, we combine the two problems. For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy. Our algorithm is trained on full-resolution data that are under-sampled retrospectively, yielding a sub-sampling pattern and reconstruction model that are customized to the type of images represented in the data set. The proposed method, which we call LOUPE (Learning-based Optimization of the Under-sampling PattErn), was implemented by modifying a U-Net, a widely-used convolutional neural network architecture, that we append with the forward model that encodes the under-sampling process. Our experiments with T1- weighted structural brain MRI scans, PD and PDFS weighted knee MRI scans show that the optimized sub-sampling pattern can yield significantly more accurate reconstructions compared to standard random uniform, variable density or cartesian under-sampling schemes. The code is made available at: https: //github.com/cagladbahadir/LOUPE ."],"dc:description.collection":["Cornell Theses and Dissertations"],"dc:format.mimetype":["application/pdf"],"dc:identifier.bibid":["11050366"],"dc:identifier.doi":["https://doi.org/10.7298/q2y0-rn35"],"dc:identifier.other":["ProQuest Submission ID: 10489","ProQuest Publication ID: 13864844"],"dc:identifier.uri":["https://hdl.handle.net/1813/67384"],"dc:language.iso":["en_US"],"dc:subject":["Computer engineering","Electrical engineering","neural networks","Optimization","Magnetic Resonance Imaging","Biomedical engineering","Compressed Sensing","Under-sampling","machine learning"],"dcterms:license":["https://hdl.handle.net/1813/59810"],"dc:title":["Learning-Based Optimization of the Under-Sampling Pattern in MRI"],"dc:type":["dissertation or thesis"],"dspace:entity.type":["Publication"],"others:access_status":["open.access"],"thesis:degree_discipline":["Biomedical Engineering"],"thesis:degree_level":["Master of Science"],"thesis:degree_name":["M.S., Biomedical Engineering"],"thesis:institution_name":["Cornell University"]},"updated_at":"2026-11-19T03:20:35Z"}