{"id":{"repo_id":"aachen","oai_identifier":"oai:publications.rwth-aachen.de:62154"},"canonical_url":"https://search.dev.ndltd.org/etd/aachen/oai:publications.rwth-aachen.de:62154","repository":{"repo_id":"aachen","name":"RWTH Aachen University","base_url":"https://publications.rwth-aachen.de/oai2d"},"display":{"title":"Vollautomatische Segmentierung von lateralen Wirbelsäulenröntgenogrammen : Auswertung und Analyse","abstract":"This study introduces a knowledge-based shape model for fully automatic segmentation of structures on radiographs, here vertebrae of the lumbar spine. The goals of this study were to examine whether the method is exact enough to determine medical structures on radiographs and to decide which practical use it may have. Additionally angles, lengths and distances were automatically calculated to examine if the algorithm is able to automatically assess functional measurements. Special focus was set on the different quality of radiographs. Therefore quality standards were created to divide radiographs into three different categories due to their difficulty for an automatic segmentation. To evaluate the accuracy of the method the automatically segmented structures were compared with manually drawn outlines according to different benchmarks. Particularly the organs outlines and outstanding structure points were verified. The automatically segmented corner points were used to calculate the intervertebral angle, the height of every single vertebra, the height of the intervertebral disc and the sagittal displacement. These figures were compared to manually identified references. Finally the algorithm was transferred from lumbar to cervical spine. The study achieved the following basic results: As the method relies on a statistical model a perfect segmentation of an individual structure cannot be achieved. The method is appropriate to distinguish the structures it was trained on. Outlines were recognized with a certain precision to define from neighbouring structures. Significant structure points were identified with minor deviation. The accuracy of the method decreases with declining image quality. Being a fully automatic method results are always reproducible. But this also means that a primary false estimation of the algorithm necessarily leads to false results which cannot be amended automatically. Functional measurements can be estimated automatically by the computer. It is able calculate both angles and distances. Trained for different closed structures the algorithm may be installed for many different tasks. The algorithm disposes of both a topological order and certain shape information which enable it to ignore interferences in the radiographs. This leads to acceptable results even on images with poor quality. On the basis of the results of this study the introduced algorithm may - with certain restrictions – assist in clinical routine.","abstract_html":"This study introduces a knowledge-based shape model for fully automatic segmentation of structures on radiographs, here vertebrae of the lumbar spine. The goals of this study were to examine whether the method is exact enough to determine medical structures on radiographs and to decide which practical use it may have. Additionally angles, lengths and distances were automatically calculated to examine if the algorithm is able to automatically assess functional measurements. Special focus was set on the different quality of radiographs. Therefore quality standards were created to divide radiographs into three different categories due to their difficulty for an automatic segmentation. To evaluate the accuracy of the method the automatically segmented structures were compared with manually drawn outlines according to different benchmarks. Particularly the organs outlines and outstanding structure points were verified. The automatically segmented corner points were used to calculate the intervertebral angle, the height of every single vertebra, the height of the intervertebral disc and the sagittal displacement. These figures were compared to manually identified references. Finally the algorithm was transferred from lumbar to cervical spine. The study achieved the following basic results: As the method relies on a statistical model a perfect segmentation of an individual structure cannot be achieved. The method is appropriate to distinguish the structures it was trained on. Outlines were recognized with a certain precision to define from neighbouring structures. Significant structure points were identified with minor deviation. The accuracy of the method decreases with declining image quality. Being a fully automatic method results are always reproducible. But this also means that a primary false estimation of the algorithm necessarily leads to false results which cannot be amended automatically. Functional measurements can be estimated automatically by the computer. It is able calculate both angles and distances. Trained for different closed structures the algorithm may be installed for many different tasks. The algorithm disposes of both a topological order and certain shape information which enable it to ignore interferences in the radiographs. This leads to acceptable results even on images with poor quality. On the basis of the results of this study the introduced algorithm may - with certain restrictions – assist in clinical routine.","abstract_has_math":false,"creators":["Brandt, Alexander Sascha"],"institution":"Publikationsserver der RWTH Aachen University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Wein, Berthold B."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2005,"date_issued":"2005","date_published":"2005","updated_at":"2026-07-30T19:43:19Z","subjects":["info:eu-repo/classification/ddc/610","Medizin","Segmentierung","Active Shape Model","Wirbelsäule"],"languages":["ger"],"rights":["info:eu-repo/semantics/openAccess"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-123744%22"],"render_values":[{"text":"https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-123744%22","href":"https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-123744%22","code":true}]}]},"links":{"outbound_url":"https://publications.rwth-aachen.de/record/62154","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Wein, Berthold B."]},{"key":"dc:creator","label":"Author","values":["Brandt, Alexander Sascha"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:coverage","label":"Dc Coverage","values":["DE"]},{"key":"dc:date","label":"Dc Date","values":["2005"]},{"key":"dc:publisher","label":"Institution","values":["Publikationsserver der RWTH Aachen University"]},{"key":"dc:relation","label":"Dc Relation","values":["info:eu-repo/semantics/altIdentifier/urn/urn:nbn:de:hbz:82-opus-12190"]},{"key":"dc:type","label":"Dc Type","values":["info:eu-repo/semantics/doctoralThesis","info:eu-repo/semantics/publishedVersion"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["info:eu-repo/classification/ddc/610","Medizin","Segmentierung","Active Shape Model","Wirbelsäule"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["ger"]},{"key":"dc:rights","label":"Dc Rights","values":["info:eu-repo/semantics/openAccess"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://publications.rwth-aachen.de/record/62154","https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-123744%22"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["This study introduces a knowledge-based shape model for fully automatic segmentation of structures on radiographs, here vertebrae of the lumbar spine. The goals of this study were to examine whether the method is exact enough to determine medical structures on radiographs and to decide which practical use it may have. Additionally angles, lengths and distances were automatically calculated to examine if the algorithm is able to automatically assess functional measurements. Special focus was set on the different quality of radiographs. Therefore quality standards were created to divide radiographs into three different categories due to their difficulty for an automatic segmentation. To evaluate the accuracy of the method the automatically segmented structures were compared with manually drawn outlines according to different benchmarks. Particularly the organs outlines and outstanding structure points were verified. The automatically segmented corner points were used to calculate the intervertebral angle, the height of every single vertebra, the height of the intervertebral disc and the sagittal displacement. These figures were compared to manually identified references. Finally the algorithm was transferred from lumbar to cervical spine. The study achieved the following basic results: As the method relies on a statistical model a perfect segmentation of an individual structure cannot be achieved. The method is appropriate to distinguish the structures it was trained on. Outlines were recognized with a certain precision to define from neighbouring structures. Significant structure points were identified with minor deviation. The accuracy of the method decreases with declining image quality. Being a fully automatic method results are always reproducible. But this also means that a primary false estimation of the algorithm necessarily leads to false results which cannot be amended automatically. Functional measurements can be estimated automatically by the computer. It is able calculate both angles and distances. Trained for different closed structures the algorithm may be installed for many different tasks. The algorithm disposes of both a topological order and certain shape information which enable it to ignore interferences in the radiographs. This leads to acceptable results even on images with poor quality. On the basis of the results of this study the introduced algorithm may - with certain restrictions – assist in clinical routine."]},{"key":"dc:source","label":"Dc Source","values":["Aachen : Publikationsserver der RWTH Aachen University VII, 130 S. : Ill., graph. Darst. (2005). = Aachen, Techn. Hochsch., Diss., 2005"]},{"key":"dc:title","label":"Title","values":["Vollautomatische Segmentierung von lateralen Wirbelsäulenröntgenogrammen : Auswertung und Analyse"]}]}],"canonical_facts":{"dc:contributor":["Wein, Berthold B."],"dc:coverage":["DE"],"dc:creator":["Brandt, Alexander Sascha"],"dc:date":["2005"],"dc:description":["This study introduces a knowledge-based shape model for fully automatic segmentation of structures on radiographs, here vertebrae of the lumbar spine. The goals of this study were to examine whether the method is exact enough to determine medical structures on radiographs and to decide which practical use it may have. Additionally angles, lengths and distances were automatically calculated to examine if the algorithm is able to automatically assess functional measurements. Special focus was set on the different quality of radiographs. Therefore quality standards were created to divide radiographs into three different categories due to their difficulty for an automatic segmentation. To evaluate the accuracy of the method the automatically segmented structures were compared with manually drawn outlines according to different benchmarks. Particularly the organs outlines and outstanding structure points were verified. The automatically segmented corner points were used to calculate the intervertebral angle, the height of every single vertebra, the height of the intervertebral disc and the sagittal displacement. These figures were compared to manually identified references. Finally the algorithm was transferred from lumbar to cervical spine. The study achieved the following basic results: As the method relies on a statistical model a perfect segmentation of an individual structure cannot be achieved. The method is appropriate to distinguish the structures it was trained on. Outlines were recognized with a certain precision to define from neighbouring structures. Significant structure points were identified with minor deviation. The accuracy of the method decreases with declining image quality. Being a fully automatic method results are always reproducible. But this also means that a primary false estimation of the algorithm necessarily leads to false results which cannot be amended automatically. Functional measurements can be estimated automatically by the computer. It is able calculate both angles and distances. Trained for different closed structures the algorithm may be installed for many different tasks. The algorithm disposes of both a topological order and certain shape information which enable it to ignore interferences in the radiographs. This leads to acceptable results even on images with poor quality. On the basis of the results of this study the introduced algorithm may - with certain restrictions – assist in clinical routine."],"dc:identifier":["https://publications.rwth-aachen.de/record/62154","https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-123744%22"],"dc:language":["ger"],"dc:publisher":["Publikationsserver der RWTH Aachen University"],"dc:relation":["info:eu-repo/semantics/altIdentifier/urn/urn:nbn:de:hbz:82-opus-12190"],"dc:rights":["info:eu-repo/semantics/openAccess"],"dc:source":["Aachen : Publikationsserver der RWTH Aachen University VII, 130 S. : Ill., graph. Darst. (2005). = Aachen, Techn. Hochsch., Diss., 2005"],"dc:subject":["info:eu-repo/classification/ddc/610","Medizin","Segmentierung","Active Shape Model","Wirbelsäule"],"dc:title":["Vollautomatische Segmentierung von lateralen Wirbelsäulenröntgenogrammen : Auswertung und Analyse"],"dc:type":["info:eu-repo/semantics/doctoralThesis","info:eu-repo/semantics/publishedVersion"]},"updated_at":"2026-07-30T19:43:19Z"}