Robust segmentation of human cardiac contours from spatial magnetic resonance images [Elektronische Ressource] / Haythem El- Messiry

Abteilung NeuroinformatikProf. Dr. Gun ther PalmRobust Segmentation of Human Cardiac Contours fromSpatial Magnetic Resonance ImagesDissertation zur Erlangung des DoktogradesDr.rer.nat.der Fakult at fur Informatik der Universit at UlmHaythem El-Messiryaus Agypten2004Amtierender Dekan: Prof. Dr. Friedrich von Henke1. Gutachter: Prof. Dr. Heiko Neumann2. Gutachter: Prof. Dr. Gun ther PalmTag der Promotion:AbstractAutomated segmentation to nd the endocardial boundary of the left heart ventriclefrom magnetic resonance (MR) images has shown to be a di cult task. One of themajor problems related to the detection of the boundary are the shortcomings typicalof discrete data, such as sampling artifacts and noise, which may cause the shapeboundaries to be indistinct and disconnected. Furthermore, the structures insidethe ventricular cavities, such as papillary muscles, are often indistinguishable fromstructures of interest for diagnostic analysis, such as the moving inner heart boundary.Thus, segmentation is error-prone and often incomplete.The aim of this work is to develop a model towards an automatic segmentationof the endocardial border. The proposed method is composed of two phases: Thesegmentation phase uses a bottom-up multi-scale analysis, based mainly on morpho-logical scale-space processing by decomposing the image into a number of scales ofdi eren t structure size.