The use of affine image registration based on normalized mutual information (NMI) has recently been proposed by Frangi et al. as an automatic method for assessing brachial artery flow mediated dilation (FMD) for the characterization of endothelial function. Even though this method solves many problems of previous approaches, there are still some situations that can lead to misregistration between frames, such as the presence of adjacent vessels due to probe movement, muscle fibres or poor image quality. Despite its widespread use as a registration metric and its promising results, MI is not the panacea and can occasionally fail. Previous work has attempted to include spatial information into the image similarity metric. Among these methods the direct estimation of α-MI through Minimum Euclidean Graphs allows to include spatial information and it seems suitable to tackle the registration problem in vascular images, where well oriented structures corresponding to vessel walls and muscle fibres are present. The purpose of this work is twofold. Firstly, we aim to evaluate the effect of including spatial information in the performance of the method suggested by Frangi et al. by using α-MI of spatial features as similarity metric. Secondly, the application of image registration to long image sequences in which both rigid motion and deformation are present will be used as a benchmark to prove the value of α-MI as a similarity metric, and will also allow us to make a comparative study with respect to NMI.
Clinical studies report that impaired endothelial function is associated with Cardio-Vascular Diseases (CVD) and their risk factors. One commonly used mean for assessing endothelial function is Flow-Mediated Dilation (FMD). Classically, FMD is quantified using local indexes e.g. maximum peak dilation. Although such parameters have been successfully linked to CVD risk factors and other clinical variables, this description does not consider all the information contained in the complete vasodilation curve. Moreover, the relation between flow impulse and the vessel vasodilation response to this stimulus, although not clearly known, seems to be important and is not taken into account in the majority of studies. In this paper we propose a novel global parameterization for the vasodilation and the flow curves of a FMD test. This parameterization uses Principal Component Analysis (PCA) to describe independently and jointly the variability of flow and FMD curves. These curves are obtained using computerized techniques (based on edge detection and image registration, respectively) to analyze the ultrasound image sequences. The global description obtained through PCA yields a detailed characterization of the morphology of such curves allowing the extraction of intuitive quantitative information of the vasodilation process and its interplay with flow changes. This parameterization is consistent with traditional measurements and, in a database of 177 subjects, seems to correlate more strongly (and with more clinical parameters) than classical measures to CVD risk factors and clinical parameters such as LDL- and HDL-Cholesterol.