Normal and Pathological Cardiac Simulations Based on Realistic FE Models


W. Paul Segars1, Alexander I. Veress2, Jeffrey A. Weiss2, Raimond Winslow1, Michael Miller1, Benjamin M. W. Tsui1, and Grant T. Gullberg3

1Johns Hopkins University,  2University of Utah, , 3Lawrence Berkeley National Laboratory


The 4D NURBS-based Cardiac-Torso (NCAT) phantom, which provides a realistic model of the normal human anatomy and cardiac and respiratory motions, is used in medical imaging research to evaluate and improve imaging devices and techniques, especially dynamic cardiac applications. One current limitation to the phantom is that it lacks the ability to accurately simulate normal variations in the cardiac anatomy and function as well as abnormal anatomical variations or functions that result from cardiac diseases such as cardiovascular disease (CVD). The goal of this work is to greatly enhance the 4D NCAT phantom by incorporating a more physiologically based, 4D finite-element (FE) mechanical model of the functioning human heart capable of realistically simulating normal and abnormal anatomical variations and cardiac motions.


The geometry and fiber architecture of the FE mechanical model will be based on state-of-the-art high-resolution diffusion tensor MRI (DTMRI) data of a normal human subject. The myocardial wall will be represented as a transversely isotropic hyperelastic material. A time varying elastance model will be used to simulate fiber contraction, and physiological intraventricular systolic pressure-time curves will be applied to simulate the motion over the cardiac cycle. We will use methods from computational anatomy to define a statistical formulation for normal and abnormal variations in the cardiac anatomy and fiber architecture. In computational anatomy, populations of anatomies are studied by mapping target anatomies to a common template anatomy. Statistics can be collected of the transforms required to match a template to many different targets and converted into distributions to mathematically model the anatomical variations of a population. Using these techniques, the statistical formulations for the normal and abnormal cardiac variations will be derived based on an analysis of several sets of high-resolution multi-slice CT (MSCT) data and tagged MRI data from anatomically diverse normal and abnormal (CVD) patients. In addition to modeling variations in the cardiac geometry, the material properties of the mechanical heart model will be altered, based on the literature, to model variations in the normal and abnormal function of the heart. Using these methods, the base FE heart model can be transformed into anatomically variable models representative of a patient population with normal and abnormal cardiac function. The methods described above will be validated through comparisons of cardiac function predicted from anatomically diverse 4D FE models to that observed in tagged MRI and echocardiography patient data.


The 4D FE cardiac model, when combined with the 4D NCAT phantom, will provide realistic, predictive imaging data of a patient population with normal and abnormal anatomical variations and functions of the heart. With this ability, the resulting enhanced 4D NCAT phantom will have a broad impact in the study and diagnosis of cardiovascular disease especially using medical imaging techniques such as MRI, x-ray CT and emission computed tomography including SPECT and PET. It will provide a vital simulation tool for the understanding of the underlying mechanisms of CVD and its effect on the normal functioning of the heart and for evaluating and improving existing and emerging 4D imaging techniques used in its diagnosis.


Subject of NIH Application R01HL082843