{"id":{"repo_id":"cornell","oai_identifier":"oai:ecommons.cornell.edu:1813/67478"},"canonical_url":"https://search.ndltd.org/etd/cornell/oai:ecommons.cornell.edu:1813/67478","repository":{"repo_id":"cornell","name":"Cornell University","base_url":"https://ecommons.cornell.edu/server/oai/request"},"display":{"title":"Inverse Problem In Quantitative Susceptibility Mapping: Numerical And Machine Learning Approaches","abstract":"Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework.","abstract_html":"Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework.","abstract_has_math":false,"creators":["Liu, Zhe"],"institution":"Cornell University","degree_name":"Ph.D., Biomedical Engineering","degree_level":"Doctor of Philosophy","degree_discipline":"Biomedical Engineering","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":["Wang, Yi"],"committee_members":["Doerschuk, Peter"],"year":2019,"date_issued":"2019-05-30","date_published":"2019-05-30","updated_at":"2026-11-19T03:20:50Z","subjects":["Medical imaging","Mathematics","MRI","QSM","Computer science","preconditioning","machine learning"],"languages":["en_US"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/k08a-nm52"],"render_values":[{"text":"https://doi.org/10.7298/k08a-nm52","href":"https://doi.org/10.7298/k08a-nm52","code":true}]},{"key":"dc:identifier.bibid","label":"Dc Identifier Bibid","values":["11050460"],"render_values":[{"text":"11050460","href":null,"code":true}]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 11290","ProQuest Publication ID: 13428363"],"render_values":[{"text":"ProQuest Submission ID: 11290","href":null,"code":true},{"text":"ProQuest Publication ID: 13428363","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1813/67478","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/1813/67478","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%2F67478","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.chair","label":"Chair","values":["Wang, Yi"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Doerschuk, Peter"]},{"key":"dc:creator","label":"Author","values":["Liu, Zhe"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2019-10-15T15:32:54Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2019-10-15T15:32:54Z"]},{"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":["Doctor of Philosophy"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D., 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":["Medical imaging","Mathematics","MRI","QSM","Computer science","preconditioning","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":["11050460"]},{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/k08a-nm52"]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 11290","ProQuest Publication ID: 13428363"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1813/67478"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"cerif:openaire","label":"Cerif Openaire","values":["<Publication xmlns=\"https://www.openaire.eu/cerif-profile/1.1/\" id=\"9ecff89b-6642-4a25-bb07-1b5d3939a425\"> <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>Inverse Problem In Quantitative Susceptibility Mapping: Numerical And Machine Learning Approaches</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019-05-30</PublicationDate> <DOI>https://doi.org/10.7298/k08a-nm52</DOI> <Authors> <Author> <DisplayName>Liu, Zhe</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <Keyword>Medical imaging</Keyword> <Keyword>Mathematics</Keyword> <Keyword>MRI</Keyword> <Keyword>QSM</Keyword> <Keyword>Computer science</Keyword> <Keyword>preconditioning</Keyword> <Keyword>machine learning</Keyword> <Abstract>Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework.</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"]},{"key":"cris:virtual.author","label":"Cris Virtual Author","values":["Liu, Zhe"]},{"key":"cris:virtualsource.collection","label":"Cris Virtualsource Collection","values":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"]},{"key":"dc:description.abstract","label":"Abstract","values":["Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework."]},{"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":["Inverse Problem In Quantitative Susceptibility Mapping: Numerical And Machine Learning Approaches"]},{"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=\"9ecff89b-6642-4a25-bb07-1b5d3939a425\"> <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>Inverse Problem In Quantitative Susceptibility Mapping: Numerical And Machine Learning Approaches</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019-05-30</PublicationDate> <DOI>https://doi.org/10.7298/k08a-nm52</DOI> <Authors> <Author> <DisplayName>Liu, Zhe</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <OrgUnit /> </Publisher> </Publishers> <Keyword>Medical imaging</Keyword> <Keyword>Mathematics</Keyword> <Keyword>MRI</Keyword> <Keyword>QSM</Keyword> <Keyword>Computer science</Keyword> <Keyword>preconditioning</Keyword> <Keyword>machine learning</Keyword> <Abstract>Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework.</Abstract> <Access xmlns=\"http://purl.org/coar/access_right\" > </Access> </Publication>"],"cris:virtual.author":["Liu, Zhe"],"cris:virtualsource.collection":["5893a6ea-7af3-41d7-abc6-04bcd26ab5df"],"dc:contributor.chair":["Wang, Yi"],"dc:contributor.committeemember":["Doerschuk, Peter"],"dc:creator":["Liu, Zhe"],"dc:date.accessioned":["2019-10-15T15:32:54Z"],"dc:date.available":["2019-10-15T15:32:54Z"],"dc:date.issued":["2019-05-30"],"dc:description.abstract":["Magnetic susceptibility reflects the concentration of bio-metal elements such as iron, calcium or gadolinium, providing means to investigate diseases such as multiple sclerosis, Alzheimer’s disease, hemorrhage and calcification. Numerous approaches have been proposed to provide magnetic susceptibility estimation from magnetic resonance imaging (MRI). While those methods are designed for specific body parts or pathologies, a unified framework is elusive from literature for reliable susceptibility estimation in both normal and pathological scenarios. This thesis developed algorithms that improve the accuracy, robustness and applicability of quantitative susceptibility mapping (QSM) for both healthy and pathological subjects. First, a dedicated regularized model was proposed to enable automated zero reference for QSM using cerebrospinal fluid. Second, convolutional neural network was combined with numerical optimization for superior anatomical contrast in QSM reconstruction. Finally, a total field inversion approach was presented to enable QSM for both healthy subject and hemorrhage patient. With the technical advances in this thesis, QSM requires less manual effort in susceptibility quantification, admits detailed recovery of anatomical structures and applies to both healthy subject and patient via a unified framework."],"dc:description.collection":["Cornell Theses and Dissertations"],"dc:format.mimetype":["application/pdf"],"dc:identifier.bibid":["11050460"],"dc:identifier.doi":["https://doi.org/10.7298/k08a-nm52"],"dc:identifier.other":["ProQuest Submission ID: 11290","ProQuest Publication ID: 13428363"],"dc:identifier.uri":["https://hdl.handle.net/1813/67478"],"dc:language.iso":["en_US"],"dc:subject":["Medical imaging","Mathematics","MRI","QSM","Computer science","preconditioning","machine learning"],"dcterms:license":["https://hdl.handle.net/1813/59810"],"dc:title":["Inverse Problem In Quantitative Susceptibility Mapping: Numerical And Machine Learning Approaches"],"dc:type":["dissertation or thesis"],"dspace:entity.type":["Publication"],"others:access_status":["open.access"],"thesis:degree_discipline":["Biomedical Engineering"],"thesis:degree_level":["Doctor of Philosophy"],"thesis:degree_name":["Ph.D., Biomedical Engineering"],"thesis:institution_name":["Cornell University"]},"updated_at":"2026-11-19T03:20:50Z"}