RAI Labs Duke
Breast CAD | Chest CAD | Lung Dis CT | XCAT-MOBY-ROBY | Breast Models | FE Cardiac Model
Quant. Image | Emerg. Quant. Imaging | Perf. Metrology | Clinical Trials | Emerg. Clinical

Multi-scale Finite-Element Cardiac Model

Integrated computational models of the heart have long been developed in order to gain a greater understanding of the normal and pathological function of the heart. A key component of these models is the myofiber architecture of the muscle tissue which dictates the heart’s electrical and mechanical functions in health and disease. Due to computational concerns and the lack of high-resolution imaging data of the human myofiber architecture, previous computational models were based on animal data and were defined for just the right and left ventricles. With the development of more efficient algorithms for electromechanical modeling as well as advances in computational power and imaging technologies such as multi-slice CT and diffusion tensor MRI, it is now feasible to develop more complex and detailed models for the human heart. The long term goal of this project is to develop and validate a 4D multi-scale finite-element (FE) computational model of the 4-chamber human heart that is capable of realistically simulating normal and abnormal cardiac anatomy and function based on state-of-

   
 

 

Figure 1. Initial FE model for the normal LV at end-diastole and end-systole.

the-art human imaging data.

The proposed heart model has enormous potential in both education and research in the areas of biomechanics, biophysics, and physiology.  It may provide a deeper understanding of the complexity of the human heart at multiple levels and establish the basis of its function in health and disease. It will provide a realistic framework to link structure and function from the cellular level to that of the intact human heart and expand the understanding of anatomical variations found in the general population. The primary application for the cardiac model will be a simulation tool for imaging research and education. When combined with a digital phantom for the human body, the model will provide realistic, predictive multi-modality patient imaging data from anatomically diverse subjects in health and disease. With this ability, the model will provide a unique and vital tool to quantitatively evaluate and compare current and emerging 4D imaging techniques used in the diagnosis of cardiovascular disease. It may also provide simulated data using various procedures and scanning parameters to train physicians.

 

   
 

 

Figure 2. Strain analysis of the normal LV model and two models simulating ischemia. The outlined region of LV in the anterior view indicates the location of the ischemia and the cutting plane indicates the location of the short axis (SA) slices. (Right columns) SA cross-sections defined at end-systole of the normal, sub-endocardial, and the transmural ischemic models. Radial, circumferential, and fiber strain results are shown for each model. The normal and diseased models produce strain results consistent with published values based on animal experiments.

   

Figure 1 shows an initial FE model we’ve developed for the left ventricle of the human heart. The FE model realistically reproduces the normal contracting, twisting motion of the heart. By manipulating the material properties of this model, we can simulate diseased states of the heart. Figure 2 shows the strain analysis of the normal LV model as well as two models simulating ischemia. The first disease model was created with a sub-endocardial anterior ischemic region extending halfway through the thickness of the LV myocardium. The second model was created with a transmural ischemic region in the same location encompassing the full thickness of the LV wall. The fiber strain predictions from the normal and ischemic models compare very favorably with published values based on animal experiments. This demonstrates the great potential FE models have to realistically simulate normal and abnormal states of the heart.

 

This work will be done in collaboration with the University of Washington, the University of California San Diego, and the Johns Hopkins University.

 
Radiographic Tech. | Low-Cost Tomo | Optimiz. of Rad. Ther. | Personal CAD | Adapt. Educ.