Statistical Atlases and Computational Models of the Heart

Recently, there has been considerable progress in the construction and appli- tion of cardiac atlases and computational models which integrate heart shape, function, and physiology. Severalmajorinitiatives haveidenti?ed computational and morphological atlases as a major infrastructural platform, for instance the Physiome project and the European Virtual PhysiologicalHuman project. N- invasive cardiovascular imaging plays an important role in de?ning the com- tational domain, the boundary/initial conditions, and tissue function and pr- erties. Hence, one of the most important current challenges in the ?eld is the development of robust and e?ective methods for the parameterization and p- sonalizationofthesecomputationalmodelsusingonlyminimally-invasiveclinical imaging. However,in orderto evaluatethe model outputandachieveclinical- pact, such personalized models have to be both augmented with and compared to generic knowledge on the healthy and pathological heart. This knowledge can be acquired through the building of statistical models of the heart. Several e?orts are now established to provide web-accessible structural and functional atlases of the normal and pathological heart for clinical, research, and edu- tionalpurposes. Webelievealltheseapproacheswillonlybee?ectivelydeveloped throughcollaborationacrossthe full researchscopeof the imaging andmodeling communities. Integrative models of cardiac function are important for understanding d- ease,evaluating treatment,and planning intervention. To providea focus for the developingarrayoftechniques whichunderpin the applicationofthese models in the clinic a simulation challenge was included in the workshop. The goal of this challenge was to compare strategies for the personaliszation of di?erent cardiac computationalmodelswith experimentaldata. A completedatasetwasprovided in advance, containing the cardiac geometry and ?bre orientations from MRI as well as epicardial transmembrane potentials from optical mapping.