Constrained segmentation of cardiac MR image sequences

Üzümcü, M. (2007)

Doctoral Thesis

Cardiovascular diseases are highly prevalent in the western world. With the aging of the population, the number of people suffering from CVD is still increasing. Therefore, the amount of diagnostic assessments and thus, the number of image acquisitions will increase accordingly. Considering the high quantities of data produced by 3D and 4D imaging modalities, such as CT and MRI, manual contouring and diagnosis currently require much time and attention of the radiologists and cardiologists. This justifies the need for (semi-) automatic segmentation methods; developing these as the focus of the research presented in this thesis.

• Knowledge based segmentation
• Cardiac
• Dynamic programming
• Active shape models
• Active appearance models
• Independent Component Analysis
• Principle Component Analysis
• Statistical Shape Analysis

• All authors: Üzümcü, M.
• Supervisor: Reiber, J.H.C.
• Co-supervisor: Lelieveldt, B.P.F.
• Qualification: Doctor (dr.)
• Awarding Institution: Leiden University Medical Center (LUMC), Leiden University
• Date: 2007-09-12
• ISBN (print): 9789090221861
• Sponsorship: Nucletron B.V. Medis medical imaging systems B.V. Stichting Beeldverwerking Foundation Imago Oegstgeest