The FIMH is the International Conference on Functional Imaging and Modeling of the Heart. This biennial scientific event aims to integrate the research and development efforts in the fields of cardiovascular modeling and image analysis. The main goal is to encourage collaboration among scientists in signal and image processing, imaging, applied mathematics, biophysics, biomedical engineering, and experts in cardiology, radiology, biology and physiology.
Characterization of Myocardial Velocities by Multiple Kernel Learning: Application to Heart Failure with Preserved Ejection Fraction
Sergio Sanchez-Martinez, Nicolas Duchateau, Bart Bijnens, Tamás Erdei, Alan Fraser, and Gemma Piella
Abstract: This study aims at improving the characterization of myocardial velocities in the context of heart failure with preserved ejection fraction (HFPEF) by combining multiple descriptors. Using a multiple kernel learning (MKL) technique, it allows the combination of data of different natures towards the learning. The methodology is applied to 2D sequences from a stress echocardiography protocol from 33 subjects (21 healthy controls and 12 HFPEF subjects). The method provides a novel way to tackle the understanding of the HFPEF syndrome, in contrast with the diagnostic issues surrounding it in the current clinical practice. Notably, the results confirm that the characterization of the myocardial functional response to stress in this syndrome is improved by the joint analysis of multiple relevant features.
Subject Independent Reference Frame for the Left Ventricular Detailed Cardiac Anatomy
Bruno Paun, Bart Bijnens, and Constantine Butakoff
Abstract: Mapping of surfaces to a parametric domain is a widely used tool in medical imaging for analysis and localization of injured tissue. By assigning the same coordinate values to specific anatomical landmarks, parametrization allows us putting into correspondence surfaces of anatomical shapes with inherently different geometry and facilitates integration of data acquired by different imaging modalities. In this paper we propose a method for subject independent anatomical parametrization of the left ventricular (LV) wall that includes trabeculations, papillary muscles and false tendons. The method relies on a disk parametrization of the LV smooth epicardium and mapping the interior of the ventricular cavity using ray casting. In this way we define a common reference frame whereupon any LV is mapped in a consistent way thus allowing for statistical analysis and comparisons between different patients.