{"id":{"repo_id":"aachen","oai_identifier":"oai:publications.rwth-aachen.de:56903"},"canonical_url":"https://search.dev.ndltd.org/etd/aachen/oai:publications.rwth-aachen.de:56903","repository":{"repo_id":"aachen","name":"RWTH Aachen University","base_url":"https://publications.rwth-aachen.de/oai2d"},"display":{"title":"Höherdimensionale Modelle zur Segmentierung biologischer Strukturen","abstract":"Many tasks in medical image processing require the robust segmentation of images. Information on the position and contour of objects allows the subsequent extraction of relevant quantitative information. This task is difficult due to actual imaging modalities that provide multi-dimensional (volumetric, time-variable, and multichannel) images. A newly formulated model is able to segment objects with arbitrary occurrence in images of any dimension. Model based segmentation methods are categorized. Subsequently, it is possible to formulate specifications that a model must meet for the robust segmentation of medical images. According to these specifications, a balloon-model is introduced. Objects are represented by a simplicial complex. Using mechanic simulations, this model is deformed to adapt to significant structures in an image. For the computation of image influences in single- and multichannel images, subsets of the same dimension as the image space itself are taken into account. The balloon-model is combined with a shape-based model. Shape knowledge from an automatically generated point distribution model is used to compute directed shape forces. The combination of all forces results in a segmentation result even if an initial contour is not given. The intersection of simplexes forms an inconsistency of the contour. This frequent problem for active contours is solved by methods that detect and correct such intersections. If necessary, these methods adaptively change the topology of objects. Further methods were developed to allow the transfer into clinical routine. The required parameter setting can be trained based on an exemplary segmentation. For heterogeneous image sets, more than one exemplary segmentation can be given. Then, an individual parameter set is computed for each image using global texture features and their similarity to prototype images. Non-contextual experiments on synthetic image material quantify the quality of segmentations for varying image properties and the dependency of the model on parameter choices. For contextual tests on medical images, usually no valid reference segmentation is known. Therefore, a silver-standard method to create synthetic images with realistic textures and contours was developed. The model was exemplary applied to immunohistochemically stained micrographs of neurons, CTs of vertebrae following prolaps of intervertebral discs, a MR of the beating heart, and laryngoscopic color video sequences. The robustness of segmentations was quantified in all applications.","abstract_html":"Many tasks in medical image processing require the robust segmentation of images. Information on the position and contour of objects allows the subsequent extraction of relevant quantitative information. This task is difficult due to actual imaging modalities that provide multi-dimensional (volumetric, time-variable, and multichannel) images. A newly formulated model is able to segment objects with arbitrary occurrence in images of any dimension. Model based segmentation methods are categorized. Subsequently, it is possible to formulate specifications that a model must meet for the robust segmentation of medical images. According to these specifications, a balloon-model is introduced. Objects are represented by a simplicial complex. Using mechanic simulations, this model is deformed to adapt to significant structures in an image. For the computation of image influences in single- and multichannel images, subsets of the same dimension as the image space itself are taken into account. The balloon-model is combined with a shape-based model. Shape knowledge from an automatically generated point distribution model is used to compute directed shape forces. The combination of all forces results in a segmentation result even if an initial contour is not given. The intersection of simplexes forms an inconsistency of the contour. This frequent problem for active contours is solved by methods that detect and correct such intersections. If necessary, these methods adaptively change the topology of objects. Further methods were developed to allow the transfer into clinical routine. The required parameter setting can be trained based on an exemplary segmentation. For heterogeneous image sets, more than one exemplary segmentation can be given. Then, an individual parameter set is computed for each image using global texture features and their similarity to prototype images. Non-contextual experiments on synthetic image material quantify the quality of segmentations for varying image properties and the dependency of the model on parameter choices. For contextual tests on medical images, usually no valid reference segmentation is known. Therefore, a silver-standard method to create synthetic images with realistic textures and contours was developed. The model was exemplary applied to immunohistochemically stained micrographs of neurons, CTs of vertebrae following prolaps of intervertebral discs, a MR of the beating heart, and laryngoscopic color video sequences. The robustness of segmentations was quantified in all applications.","abstract_has_math":false,"creators":["Bredno, Jörg"],"institution":"Publikationsserver der RWTH Aachen University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Oberschelp, Walter"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2002,"date_issued":"2002","date_published":"2002","updated_at":"2026-07-30T19:42:01Z","subjects":["info:eu-repo/classification/ddc/570","Biowissenschaften, Biologie"],"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-118979%22"],"render_values":[{"text":"https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-118979%22","href":"https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-118979%22","code":true}]}]},"links":{"outbound_url":"https://publications.rwth-aachen.de/record/56903","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Oberschelp, Walter"]},{"key":"dc:creator","label":"Author","values":["Bredno, Jörg"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:coverage","label":"Dc Coverage","values":["DE"]},{"key":"dc:date","label":"Dc Date","values":["2002"]},{"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-3111"]},{"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/570","Biowissenschaften, Biologie"]}]},{"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/56903","https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-118979%22"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Many tasks in medical image processing require the robust segmentation of images. Information on the position and contour of objects allows the subsequent extraction of relevant quantitative information. This task is difficult due to actual imaging modalities that provide multi-dimensional (volumetric, time-variable, and multichannel) images. A newly formulated model is able to segment objects with arbitrary occurrence in images of any dimension. Model based segmentation methods are categorized. Subsequently, it is possible to formulate specifications that a model must meet for the robust segmentation of medical images. According to these specifications, a balloon-model is introduced. Objects are represented by a simplicial complex. Using mechanic simulations, this model is deformed to adapt to significant structures in an image. For the computation of image influences in single- and multichannel images, subsets of the same dimension as the image space itself are taken into account. The balloon-model is combined with a shape-based model. Shape knowledge from an automatically generated point distribution model is used to compute directed shape forces. The combination of all forces results in a segmentation result even if an initial contour is not given. The intersection of simplexes forms an inconsistency of the contour. This frequent problem for active contours is solved by methods that detect and correct such intersections. If necessary, these methods adaptively change the topology of objects. Further methods were developed to allow the transfer into clinical routine. The required parameter setting can be trained based on an exemplary segmentation. For heterogeneous image sets, more than one exemplary segmentation can be given. Then, an individual parameter set is computed for each image using global texture features and their similarity to prototype images. Non-contextual experiments on synthetic image material quantify the quality of segmentations for varying image properties and the dependency of the model on parameter choices. For contextual tests on medical images, usually no valid reference segmentation is known. Therefore, a silver-standard method to create synthetic images with realistic textures and contours was developed. The model was exemplary applied to immunohistochemically stained micrographs of neurons, CTs of vertebrae following prolaps of intervertebral discs, a MR of the beating heart, and laryngoscopic color video sequences. The robustness of segmentations was quantified in all applications."]},{"key":"dc:source","label":"Dc Source","values":["Aachen : Publikationsserver der RWTH Aachen University VIII, 211 S. : Ill., graph. Darst. (2002). = Aachen, Techn. Hochsch., Diss., 2002"]},{"key":"dc:title","label":"Title","values":["Höherdimensionale Modelle zur Segmentierung biologischer Strukturen"]}]}],"canonical_facts":{"dc:contributor":["Oberschelp, Walter"],"dc:coverage":["DE"],"dc:creator":["Bredno, Jörg"],"dc:date":["2002"],"dc:description":["Many tasks in medical image processing require the robust segmentation of images. Information on the position and contour of objects allows the subsequent extraction of relevant quantitative information. This task is difficult due to actual imaging modalities that provide multi-dimensional (volumetric, time-variable, and multichannel) images. A newly formulated model is able to segment objects with arbitrary occurrence in images of any dimension. Model based segmentation methods are categorized. Subsequently, it is possible to formulate specifications that a model must meet for the robust segmentation of medical images. According to these specifications, a balloon-model is introduced. Objects are represented by a simplicial complex. Using mechanic simulations, this model is deformed to adapt to significant structures in an image. For the computation of image influences in single- and multichannel images, subsets of the same dimension as the image space itself are taken into account. The balloon-model is combined with a shape-based model. Shape knowledge from an automatically generated point distribution model is used to compute directed shape forces. The combination of all forces results in a segmentation result even if an initial contour is not given. The intersection of simplexes forms an inconsistency of the contour. This frequent problem for active contours is solved by methods that detect and correct such intersections. If necessary, these methods adaptively change the topology of objects. Further methods were developed to allow the transfer into clinical routine. The required parameter setting can be trained based on an exemplary segmentation. For heterogeneous image sets, more than one exemplary segmentation can be given. Then, an individual parameter set is computed for each image using global texture features and their similarity to prototype images. Non-contextual experiments on synthetic image material quantify the quality of segmentations for varying image properties and the dependency of the model on parameter choices. For contextual tests on medical images, usually no valid reference segmentation is known. Therefore, a silver-standard method to create synthetic images with realistic textures and contours was developed. The model was exemplary applied to immunohistochemically stained micrographs of neurons, CTs of vertebrae following prolaps of intervertebral discs, a MR of the beating heart, and laryngoscopic color video sequences. The robustness of segmentations was quantified in all applications."],"dc:identifier":["https://publications.rwth-aachen.de/record/56903","https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-118979%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-3111"],"dc:rights":["info:eu-repo/semantics/openAccess"],"dc:source":["Aachen : Publikationsserver der RWTH Aachen University VIII, 211 S. : Ill., graph. Darst. (2002). = Aachen, Techn. Hochsch., Diss., 2002"],"dc:subject":["info:eu-repo/classification/ddc/570","Biowissenschaften, Biologie"],"dc:title":["Höherdimensionale Modelle zur Segmentierung biologischer Strukturen"],"dc:type":["info:eu-repo/semantics/doctoralThesis","info:eu-repo/semantics/publishedVersion"]},"updated_at":"2026-07-30T19:42:01Z"}