A Comprehensive Approach to the Analysis of Contrast Enhanced Cardiac MR Images

Abstract

Current magnetic resonance imaging (MRI) technology allows the determination of patient-individual coronary tree structure, detection of infarctions, and assessment of myocardial perfusion. Joint inspection of these three aspects yields valuable information for therapy planning, e.g., through classification of myocardium into healthy tissue, regions showing a reversible hypoperfusion, and infarction with additional information on the corresponding supplying artery. Standard imaging protocols normally provide image data with different orientations, resolutions and coverages for each of the three aspects, which makes a direct comparison of analysis results difficult. The purpose of this work is to develop methods for the alignment and combined analysis of these images. The proposed approach is applied to 21 datasets of healthy and diseased patients from the clinical routine. The evaluation shows that, despite limitations due to typical MRI artifacts, combined inspection is feasible and can yield clinically useful information.

Automatic segmentation of pathological tissues in cardiac MRI

“…To improve this, Hsu alternates thresholding and feature analysis [4], assuming that abnormal tissues are mainly sub-endocardial and reasonably large. Heiberg introduces a level set algorithm to regularize thresholding [5], while in Hennemuth's recent approach, the threshold resulting of image intensity analysis is used to initialize a watershed segmentation [6]. However, besides the spatial coherence brought by all these methods, it is always necessary to use a connectivity-based post-processing to re ne the results.…”

“…However, besides the spatial coherence brought by all these methods, it is always necessary to use a connectivity-based post-processing to re ne the results. In this paper, we propose a new method with two main contributions: (1) the generalization of the intensity analysis presented in [6] and (2) an original variational segmentation method, the Fast Region Competition, leading to accurate results without any assumption concerning the location of abnormal tissues nor need for post-processing.…”

Phase-based non-rigid registration of myocardial perfusion MRI image sequences

“…The special problem of aligning myocardial perfusion MRI sequences has also been addressed before, overviews can be found in [1,2,3]. As noted by Milles et al [2], the proposed methods are limited when dealing with the contrast variations in the image sequences.…”

“…Clearly, an image similarity measure based on voxel differences or correlation is suboptimal in the presence of contrast variations. Alternative features for registration include, for example, landmarks [1] and mutual information [3]. As a basis for registration in this work, we instead propose to use local phase, which represents image features such edges and lines but is invariant to their magnitude.…”

Developing advanced mathematical models for detecting abnormalities in 2D/3D medical structures.

“…Heiberg et al [36] augmented the intensity thresholding with a level-set-based regulation to exclude small noisy regions. Hennemuth et al [37] used the image intensity profile to initiate a watershed-based segmentation. Connected component analysis, to fill holes and exclude small noisy regions was further used to refine this segmentation.…”