Mechanical behaviour and rupture of normal and pathological human ascending aortic wall

The mechanical properties of aortic wall, both healthy and pathological, are needed in order to develop and improve diagnostic and interventional criteria, and for the development of mechanical models to assess arterial integrity. This study focuses on the mechanical behaviour and rupture conditions of the human ascending aorta and its relationship with age and pathologies. Fresh ascending aortic specimens harvested from 23 healthy donors, 12 patients with bicuspid aortic valve (BAV) and 14 with aneurysm were tensile-tested in vitro under physiological conditions. Tensile strength, stretch at failure and elbow stress were measured. The obtained results showed that age causes a major reduction in the mechanical parameters of healthy ascending aortic tissue, and that no significant differences are found between the mechanical strength of aneurysmal or BAV aortic specimens and the corresponding age-matched control group. The physiological level of the stress in the circumferential direction was also computed to assess the physiological operation range of healthy and diseased ascending aortas. The mean physiological wall stress acting on pathologic aortas was found to be far from rupture, with factors of safety (defined as the ratio of tensile strength to the mean wall stress) larger than six. In contrast, the physiological operation of pathologic vessels lays in the stiff part of the response curve, losing part of its function of damping the pressure waves from the heart.

On the Mahalanobis Distance Classification Criterion for Multidimensional Normal Distributions

Many existing engineering works model the statistical characteristics of the entities under study as normal distributions. These models are eventually used for decision making, requiring in practice the definition of the classification region corresponding to the desired confidence level. Surprisingly enough, however, a great amount of computer vision works using multidimensional normal models leave unspecified or fail to establish correct confidence regions due to misconceptions on the features of Gaussian functions or to wrong analogies with the unidimensional case. The resulting regions incur in deviations that can be unacceptable in high-dimensional models. Here we provide a comprehensive derivation of the optimal confidence regions for multivariate normal distributions of arbitrary dimensionality. To this end, firstly we derive the condition for region optimality of general continuous multidimensional distributions, and then we apply it to the widespread case of the normal probability density function. The obtained results are used to analyze the confidence error incurred by previous works related to vision research, showing that deviations caused by wrong regions may turn into unacceptable as dimensionality increases. To support the theoretical analysis, a quantitative example in the context of moving object detection by means of background modeling is given.

Classical segmentation methods on novel MR imaging: a study of brain tissue segmentation of MP2RAGE vs MPRAGE

MP2RAGE has proven to be a bias-free MR acquisition with excellent contrast between grey and white matter. We investigated the ability of three state-of-the-art algorithms to automatically extract white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF) from MPRAGE and MP2RAGE images: unified Segmentation (S) in SPM82 , its extension New Segment (NS), and an in-house Expectation-Maximization Markov Random Field tissue classification3 (EM-MRF) with Graph Cut (GC) optimization4 . Our goal is to quantify the differences between MPRAGE and MP2RAGE-based brain tissue probability maps.