Matches in SemOpenAlex for { <https://semopenalex.org/work/W2247993650> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W2247993650 abstract "Studying femur morphology on a large population of computed-tomography (CT) images requires automatic methods. This thesis presents a fully automatic CT-to- model pipeline that accurately segments and models femur morphology. The pipeline is composed of a training phase, where a statistical shape model is created, and a processing phase, which segments and models cortical bone geometry and cancellous bone mineral density (BMD) distribution. Development and testing of the pipeline was carried out on a set of 262 quantitative-CT images from the Victorian Institute of Forensic Medicine (VIFM). In the training phase, corresponding regions on a training-set of 41 femoral surfaces were automatically partitioned and grouped using region-growing and mean-shift clus- tering. These regions were used to design a region-based quartic-Lagrange femur mesh, which was fitted region-by-region to manually segmented surfaces, to train the femur statistical shape model. Validation experiments showed that this region-based shape model was more accurate and correspondent than an equivalent non-regional model. Cortical bone geometry was automatically extracted and modelled in the first step of the processing phase, using the shape mode above. Active shape modelling and cortical thickness mapping were adapted and combined to mesh the inner and outer cortical surfaces. Segmented meshes were accurate to 0.9 mm root-mean-square (RMS), and cortical thickness to 0.6 mm RMS. The method achieved a success rate of 83%. Cancellous BMD images were automatically segmented and registered in the second step of the processing phase. BMD values from CT images were mapped to a reference volume by radial basis functions (RBFs), which interpolated the mapping between segmented and reference inner cortical surface meshes. Compared to conventional free- form deformation (FFD) registration, RBF registration followed by FFD led to a four- fold reduction in run time, a surface accuracy of 0.76 mm (versus 3.7 mm), and better alignment of anatomical features. Principal component analysis of registered images showed BMD variations in clinically relevant regions. The development of the CT-to-model pipeline has enabled unsupervised data col- lection from VIFM CT scans for scientific and clinical studies of femur morphology. As a general framework, minor modifications of the pipeline will also allow unsupervised data collection for other bones and other image sets." @default.
- W2247993650 created "2016-06-24" @default.
- W2247993650 creator A5054104153 @default.
- W2247993650 date "2013-01-01" @default.
- W2247993650 modified "2023-09-28" @default.
- W2247993650 title "Development of an automated system for building a large population-based statistical model of femur morphology" @default.
- W2247993650 hasPublicationYear "2013" @default.
- W2247993650 type Work @default.
- W2247993650 sameAs 2247993650 @default.
- W2247993650 citedByCount "0" @default.
- W2247993650 crossrefType "dissertation" @default.
- W2247993650 hasAuthorship W2247993650A5054104153 @default.
- W2247993650 hasConcept C105702510 @default.
- W2247993650 hasConcept C121684516 @default.
- W2247993650 hasConcept C127313418 @default.
- W2247993650 hasConcept C144024400 @default.
- W2247993650 hasConcept C149923435 @default.
- W2247993650 hasConcept C151730666 @default.
- W2247993650 hasConcept C154945302 @default.
- W2247993650 hasConcept C199360897 @default.
- W2247993650 hasConcept C2776350087 @default.
- W2247993650 hasConcept C2780554211 @default.
- W2247993650 hasConcept C2908647359 @default.
- W2247993650 hasConcept C31487907 @default.
- W2247993650 hasConcept C31972630 @default.
- W2247993650 hasConcept C41008148 @default.
- W2247993650 hasConcept C43521106 @default.
- W2247993650 hasConcept C71924100 @default.
- W2247993650 hasConceptScore W2247993650C105702510 @default.
- W2247993650 hasConceptScore W2247993650C121684516 @default.
- W2247993650 hasConceptScore W2247993650C127313418 @default.
- W2247993650 hasConceptScore W2247993650C144024400 @default.
- W2247993650 hasConceptScore W2247993650C149923435 @default.
- W2247993650 hasConceptScore W2247993650C151730666 @default.
- W2247993650 hasConceptScore W2247993650C154945302 @default.
- W2247993650 hasConceptScore W2247993650C199360897 @default.
- W2247993650 hasConceptScore W2247993650C2776350087 @default.
- W2247993650 hasConceptScore W2247993650C2780554211 @default.
- W2247993650 hasConceptScore W2247993650C2908647359 @default.
- W2247993650 hasConceptScore W2247993650C31487907 @default.
- W2247993650 hasConceptScore W2247993650C31972630 @default.
- W2247993650 hasConceptScore W2247993650C41008148 @default.
- W2247993650 hasConceptScore W2247993650C43521106 @default.
- W2247993650 hasConceptScore W2247993650C71924100 @default.
- W2247993650 hasLocation W22479936501 @default.
- W2247993650 hasOpenAccess W2247993650 @default.
- W2247993650 hasPrimaryLocation W22479936501 @default.
- W2247993650 hasRelatedWork W1490992805 @default.
- W2247993650 hasRelatedWork W1555620887 @default.
- W2247993650 hasRelatedWork W159395664 @default.
- W2247993650 hasRelatedWork W180388651 @default.
- W2247993650 hasRelatedWork W1974178494 @default.
- W2247993650 hasRelatedWork W2025885669 @default.
- W2247993650 hasRelatedWork W2106202659 @default.
- W2247993650 hasRelatedWork W2148150519 @default.
- W2247993650 hasRelatedWork W2285353635 @default.
- W2247993650 hasRelatedWork W2528134368 @default.
- W2247993650 hasRelatedWork W2593743935 @default.
- W2247993650 hasRelatedWork W2602223469 @default.
- W2247993650 hasRelatedWork W2621761332 @default.
- W2247993650 hasRelatedWork W2755937006 @default.
- W2247993650 hasRelatedWork W2823147292 @default.
- W2247993650 hasRelatedWork W2900746287 @default.
- W2247993650 hasRelatedWork W2916290123 @default.
- W2247993650 hasRelatedWork W3025930867 @default.
- W2247993650 hasRelatedWork W3095505136 @default.
- W2247993650 hasRelatedWork W46559580 @default.
- W2247993650 isParatext "false" @default.
- W2247993650 isRetracted "false" @default.
- W2247993650 magId "2247993650" @default.
- W2247993650 workType "dissertation" @default.