Logo eprints

Joint registration and segmentation of CP-BOLD MRI

Oksuz, Ilkay (2017) Joint registration and segmentation of CP-BOLD MRI. Advisor: Tsaftaris, Prof. Sotirios. Coadvisor: Ricciardi, Prof. Emiliano . pp. 170. [IMT PhD Thesis]

Oksuz_phdthesis.pdf - Published Version
Available under License Creative Commons Attribution No Derivatives.

Download (7MB) | Preview


Joint registration and segmentation of varying contrast images is a fundamental task in the field of image analysis, despite yet open. In this thesis, novel techniques for the tasks of segmentation and registration are discussed separately and jointly. Cardiac Phaseresolved Blood Oxygen-Level-Dependent (CP-BOLD) MRI is a new contrast agent- and stress-free imaging technique for the assessment of myocardial ischemia at rest. However, it introduces varying contrast in medical image analysis applications. Therefore, establishing voxel to voxel correspondences throughout the cardiac sequence, an inevitable component of statistical analysis of these images remains challenging. Furthermore, medical background and specific segmentation difficulties associated to these images are present. Alongside with the inconsistency in myocardial intensity patterns, the changes in myocardial shape due to the heart’s motion lead to low registration performance for state-of-the-art methods. The problem of low accuracy can be explained by the lack of distinguishable features in CP-BOLD and inappropriate metric definitions in current intensity-based registration and segmentation frameworks. In this thesis, sparse representations, which are defined by a discriminative dictionary learning approach, are used to improve myocardial segmentation and registration. Initially appearance information is combined with Gabor and HOG features in a dictionary learning framework to sparsely represent features in a low dimensional space. Moreover, the motion is incorporated as additional feature to establish an unsupervised segmentation framework. For registering the cardiac sequence a new similarity metric is proposed utilizing the sparse representations. Also a joint optimization scheme for dictionary learning based feature representations is proposed using the sparse coefficients and dictionary residuals. The superior performance of the dictionary-based descriptors are showcased with several experimental results.

Item Type: IMT PhD Thesis
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
PhD Course: Computer Decision and System Science
Identification Number: 10.6092/imtlucca/e-theses/238
Date Deposited: 08 Feb 2018 10:51
URI: http://e-theses.imtlucca.it/id/eprint/238

Actions (login required, only for staff repository)

View Item View Item