{"id":{"repo_id":"usf","oai_identifier":"oai:digitalcommons.usf.edu:etd-3157"},"canonical_url":"https://search.ndltd.org/etd/usf/oai:digitalcommons.usf.edu:etd-3157","repository":{"repo_id":"usf","name":"University of South Florida","base_url":"https://digitalcommons.usf.edu/do/oai/"},"display":{"title":"Temporal Registration of Mammograms by Finite Element Simulation of MR Breast Volume Deformation","abstract":"Clinically it is important to combine information provided by mammographic images from multiple views or at different times. Taking regular mammographic screening and comparing corresponding mammograms are necessary for early detection of breast cancer, which is the key to successful treatment. However, mammograms taken at different times are often obtained under different compressions, orientations or body positions. A temporal pair of mammograms may vary quite significantly due to the spatial disparities caused by the variety in acquisition environments, including the 3D position of the breast, the amount of the pressure applied, etc. Such disparities can be corrected through the process of temporal registration. We have implemented and utilized finite element models for temporal registration of digital mammography. In our work, we applied the patient specific breast model, where patients have both mammograms and MRIs available, and generic model, where only patient mammograms are available. After we applied the temporal registration algorithm, the average error among the 14 patient datasets was 3.4 plus/minus 0.86 mm for Euclidean distance and 4.3 plus/minus 0.52 mm for predicted 2D lesion position. With generic model, the average error among the 14 patient datasets using the measure of Euclidean distance between the predicted lesion position in T1 and T2 was 5.0 plus/minus 0.74 mm for Euclidean distance and 5.7 plus/minus 0.83 mm for predicted 2D lesion position. Compared with the average lesion size (10mm~40mm), this error is acceptable. With lesion correspondence, our finite element method can be used to suppress technical variations (e.g., mammogram positioning or compression) and to emphasize genuine alterations in the breast.","abstract_html":"Clinically it is important to combine information provided by mammographic images from multiple views or at different times. Taking regular mammographic screening and comparing corresponding mammograms are necessary for early detection of breast cancer, which is the key to successful treatment. However, mammograms taken at different times are often obtained under different compressions, orientations or body positions. A temporal pair of mammograms may vary quite significantly due to the spatial disparities caused by the variety in acquisition environments, including the 3D position of the breast, the amount of the pressure applied, etc. Such disparities can be corrected through the process of temporal registration. We have implemented and utilized finite element models for temporal registration of digital mammography. In our work, we applied the patient specific breast model, where patients have both mammograms and MRIs available, and generic model, where only patient mammograms are available. After we applied the temporal registration algorithm, the average error among the 14 patient datasets was 3.4 plus/minus 0.86 mm for Euclidean distance and 4.3 plus/minus 0.52 mm for predicted 2D lesion position. With generic model, the average error among the 14 patient datasets using the measure of Euclidean distance between the predicted lesion position in T1 and T2 was 5.0 plus/minus 0.74 mm for Euclidean distance and 5.7 plus/minus 0.83 mm for predicted 2D lesion position. Compared with the average lesion size (10mm~40mm), this error is acceptable. With lesion correspondence, our finite element method can be used to suppress technical variations (e.g., mammogram positioning or compression) and to emphasize genuine alterations in the breast.","abstract_has_math":false,"creators":["Qiu, Yan"],"institution":"University of South Florida","degree_name":null,"degree_level":"Dissertation","degree_discipline":"Computer Science and Engineering","degree_department":null,"school":null,"contributors":["Dmitry B. Goldgof, Ph.D.","Rangachar Kasturi, Ph.D.","Sudeep Sarkar, Ph.D.","Lihua Li, Ph.D.","Michael VanAuker, Ph.D."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2009,"date_issued":"2009-01-01T08:00:00Z","date_published":"2009-01-01T08:00:00Z","updated_at":"2026-11-19T08:34:29Z","subjects":["Breast Cancer","Non-rigid Registration","X-ray","Computer Aided Analysis","Compression Simulation","American Studies","Arts and Humanities"],"languages":["EN-US"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.usf.edu/etd/2158","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.usf.edu/etd/2158","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.usf.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.usf.edu%3Aetd-3157","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dmitry B. Goldgof, Ph.D.","Rangachar Kasturi, Ph.D.","Sudeep Sarkar, Ph.D.","Lihua Li, Ph.D.","Michael VanAuker, Ph.D."]},{"key":"dc:creator","label":"Author","values":["Qiu, Yan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2009-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science and Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation","Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of South Florida"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Breast Cancer","Non-rigid Registration","X-ray","Computer Aided Analysis","Compression Simulation","American Studies","Arts and Humanities"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["EN-US"]},{"key":"dc:rights.license","label":"Dc Rights License","values":["default"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.usf.edu/etd/2158"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Clinically it is important to combine information provided by mammographic images from multiple views or at different times. Taking regular mammographic screening and comparing corresponding mammograms are necessary for early detection of breast cancer, which is the key to successful treatment. However, mammograms taken at different times are often obtained under different compressions, orientations or body positions. A temporal pair of mammograms may vary quite significantly due to the spatial disparities caused by the variety in acquisition environments, including the 3D position of the breast, the amount of the pressure applied, etc. Such disparities can be corrected through the process of temporal registration. We have implemented and utilized finite element models for temporal registration of digital mammography. In our work, we applied the patient specific breast model, where patients have both mammograms and MRIs available, and generic model, where only patient mammograms are available. After we applied the temporal registration algorithm, the average error among the 14 patient datasets was 3.4 plus/minus 0.86 mm for Euclidean distance and 4.3 plus/minus 0.52 mm for predicted 2D lesion position. With generic model, the average error among the 14 patient datasets using the measure of Euclidean distance between the predicted lesion position in T1 and T2 was 5.0 plus/minus 0.74 mm for Euclidean distance and 5.7 plus/minus 0.83 mm for predicted 2D lesion position. Compared with the average lesion size (10mm~40mm), this error is acceptable. With lesion correspondence, our finite element method can be used to suppress technical variations (e.g., mammogram positioning or compression) and to emphasize genuine alterations in the breast."]},{"key":"dc:title","label":"Title","values":["Temporal Registration of Mammograms by Finite Element Simulation of MR Breast Volume Deformation"]}]}],"canonical_facts":{"dc:contributor":["Dmitry B. Goldgof, Ph.D.","Rangachar Kasturi, Ph.D.","Sudeep Sarkar, Ph.D.","Lihua Li, Ph.D.","Michael VanAuker, Ph.D."],"dc:creator":["Qiu, Yan"],"dc:date":["2009-01-01T08:00:00Z"],"dc:description":["Clinically it is important to combine information provided by mammographic images from multiple views or at different times. Taking regular mammographic screening and comparing corresponding mammograms are necessary for early detection of breast cancer, which is the key to successful treatment. However, mammograms taken at different times are often obtained under different compressions, orientations or body positions. A temporal pair of mammograms may vary quite significantly due to the spatial disparities caused by the variety in acquisition environments, including the 3D position of the breast, the amount of the pressure applied, etc. Such disparities can be corrected through the process of temporal registration. We have implemented and utilized finite element models for temporal registration of digital mammography. In our work, we applied the patient specific breast model, where patients have both mammograms and MRIs available, and generic model, where only patient mammograms are available. After we applied the temporal registration algorithm, the average error among the 14 patient datasets was 3.4 plus/minus 0.86 mm for Euclidean distance and 4.3 plus/minus 0.52 mm for predicted 2D lesion position. With generic model, the average error among the 14 patient datasets using the measure of Euclidean distance between the predicted lesion position in T1 and T2 was 5.0 plus/minus 0.74 mm for Euclidean distance and 5.7 plus/minus 0.83 mm for predicted 2D lesion position. Compared with the average lesion size (10mm~40mm), this error is acceptable. With lesion correspondence, our finite element method can be used to suppress technical variations (e.g., mammogram positioning or compression) and to emphasize genuine alterations in the breast."],"dc:identifier":["https://digitalcommons.usf.edu/etd/2158"],"dc:language":["EN-US"],"dc:rights.license":["default"],"dc:subject":["Breast Cancer","Non-rigid Registration","X-ray","Computer Aided Analysis","Compression Simulation","American Studies","Arts and Humanities"],"dc:title":["Temporal Registration of Mammograms by Finite Element Simulation of MR Breast Volume Deformation"],"thesis:degree_discipline":["Computer Science and Engineering"],"thesis:degree_level":["Dissertation","Ph.D."],"thesis:institution_name":["University of South Florida"]},"updated_at":"2026-11-19T08:34:29Z"}