Cardioid Cardiac Modeling Project - overview
Cardiovascular disease remains a leading cause of death in United States. Biophysical models of cardiovascular system and heart function are promising analytic tools to assist medical practice and clinical trials. Over the years, we have developed detailed multi-scale models of cardiac mechanics that can leverage data collected across several scales of physiology in clinical and pre-clinical settings. Some examples include data from in vitro experiments, multi-modality cardiac imaging, and other cardiac functional measurements.
A major research expertise of our group is 3D high resolution modeling of cardiac mechanics. Using computational models, we simulate cardiac function under different conditions including myocardial infarction and heart failure. In the future, such detailed cardiac models could prove advantageous to healthcare professionals by providing customized and data-driven care to patients with heart-related conditions.
In a heart beat, each and every heart muscle cell is electrically excited, creating a secondary influx of calcium ions into the cell that in turn activate the myofilaments to contract. The force developed by each cell is computed using an approximate ODE (ordinary differential equations)-based model of myofilaments. The sum of the forces of the contracting cells generates the force to contract the heart and to eject the blood. However, the final force calculation require more than just summing the cell forces calculated using ODEs; there are additional contributions from the passive tissue properties, the complex arrangement of muscle fibers and the pressure of the blood to be ejected.
Achieving greater detail requires high computational demands. We are developing new algorithms to execute tissue-level cardiac models at an unprecedented level of parallelism, which exploit IBM's High Performance Computing (HPC) Platforms. An example of the model for canine heart is shown to the right.
The high resolution of the model has important implications to accurately simulate myocardial infarction (MI). An ischemic region can potentially have a complex morphology, and older models that used simple shapes of infarct are not credible. To improve the accuracy of MI simulations, we refined elements in the border of the infarcted region, achieving fine resolution around the infarct border.
This 3D model is coupled to a lumped model of blood circulation in the body. This allows us to simulate all phases of the cardiac cycle: isovolumic contraction, blood ejection, isovolumic relaxation, and ventricular filling. Below, we simulated pressure-volume loops in a simple model under different conditions.
Our goal is to make a high-resolution model that is also reliable, which is why justifies our emphasis on capturing the details of cardiac contraction. Our efforts are currently directed towards creating a new model of cardiac filament that will be able to reproduce a variety of experimental observations with Cardioid.
To make feasible model optimization of cardiovascular system, we have recently developed a novel multi-fidelity strategy for model order reduction of 3-D finite element models of ventricular mechanics. The significant improvement in computational efficiency of the low order models over the finite element model counterparts allows for efficient parametric explorations and solutions of inverse optimization problems, which is crucial for the application of biophysical models in clinical settings.
Sudden cardiac arrest and arrhythmia (SCA) is a leading cause of death in the United States with about 325,000 deaths per year (almost 900 deaths per day) (heart.org). SCA results from ventricular fibrillation and is caused by irregular electrical disturbances. A variety of factors can cause SCA; one of these factors is prescription medication. In the past, FDA-approved drugs for certain conditions were pulled from the market for causing arrhythmia. Arrhythmogenic drugs came from many classes including antidepressants, antimicrobials, and even antiarrhythmic drugs. Reliable methods to assess the small but very real risk of new candidate drugs remains a major concern for pharmaceutical companies and the regulatory agencies.
In our group, we develop computational models of cardiac electrophysiology. Using these models, it is possible to directly simulate arrhythmias, as shown below.
Cardioid is a high-resolution cardiac solver that can use ultrasound, MRI, and CT data derived from real patients. The video on the left demonstrates construction of the finite element model from cryosectional images from the Visible Human Project.
The video below shows a simulation of Torsades de Pointes, a particular type of tachyarrhythmia characterized by a slow quasi-periodic envelope modulation. Two views of the same heart are shown with red – cellular electrical activation and blue – cellular electrical relaxation. Arrhythmia, which is induced by a point stimulus simulating an ectopic acticvation, forms the spiral waves at the end of the relaxation. Spiral waves with continually changing orientation generate with the dramatic changes in amplitude in the ECG.
We have also recently incorporated sensitivity analysis and uncertainty quantification approaches to our computational pipeline, which could be used for quantification and improvement of the confidence in the predicted risk and to explore the intrinsic structure of model derived metrics.