Nuclear based imaging modalities, such as SPECT and PET, are frequently used clinically to evaluate the left ventricular (LV) viability and function. Nuclear cardiac imaging software packages can now be used to obtain estimates of ejection fraction (EF), wall motion, and wall thickening. A major limitation of current technology is that EF, wall motion, and wall thickening measures, are indirect measures of left ventricular tissue contraction (systole) and dilation (diastole). A direct measure of systolic and diastolic LV function would be myocardial deformation as quantified by local wall strain. The goal of the proposed work is to develop an image driven, subject-specific mechanical model that will provide accurate strain measurements for the left ventricle without the need to measure the pressure boundary conditions. This will be accomplished through a novel expansion of Hyperelastic Warping, an image registration technology that will be able to provide physiologically realistic contraction and dilation over the entire cardiac cycle based on the deformation information documented in standard clinical nuclear images. During the course of this work, new technology will be developed that will be able to identify and quantify nuclear image uptake information which will then be used to alter the corresponding local wall function of the models. Additionally, these models will also be able to provide an estimate of myocardial wall stress. The proposed methodology will be refined using synthetic image data produced by the 4D NURBS based cardiac-torso phantom (NCAT) image data based upon a physiologically realistic forward finite element (FE) model. The primary validation of the warping models will be made through the analysis of 6 human PET and 6 SPECT image data sets that will have corresponding tagged MRI images. The warping deformations will be compared with those determined by Harmonic Phase MRI analyses of the tagged MRI image data sets. In order to evaluate the robustness of the warping stress predictions, a series of sensitivity warping studies will be conducted using the FE/NCAT produced PET and SPECT image data sets. These analyses will be made for cases where the material parameters and fiber distributions used to define the models have been systematically altered. The warping stress predictions will be compared with those forward FE model upon which the 4D NCAT model images will be based upon. This proposal is designed to address the basic scientific and clinical needs for greater functional information through the evaluation of a measurement system that will be able to determine both systolic and diastolic regional deformation directly from nuclear images as well as providing an estimate of myocardial wall stress.
The work within this proposal will provide new technology to address the basic science and clinical need for greater functional information from non-invasive imaging. This will be accomplished through the development and validation of a measurement system which will function as an image driven mechanical model of the LV. The technology will be able to determine both systolic and diastolic regional deformation directly from nuclear images (PET and SPECT) with the potential to provide a stress estimate of the myocardium.
| Veress, Alexander I; Fung, George S K; Lee, Taek-Soo et al. (2015) The direct incorporation of perfusion defect information to define ischemia and infarction in a finite element model of the left ventricle. J Biomech Eng 137:051004 |
| Veress, Alexander I; Klein, Gregory; Gullberg, Grant T (2013) A Comparison of Hyperelastic Warping of PET Images with Tagged MRI for the Analysis of Cardiac Deformation. Int J Biomed Imaging 2013:728624 |
| Veress, A I; Raymond, G M; Gullberg, G T et al. (2013) Left ventricular finite element model bounded by a systemic circulation model. J Biomech Eng 135:54502 |
