Research

Systemic Physiology Modeling

The behaviours of the cardiovascular system can be approximated via simplified circuit representations. Using this approach, we describe the effects of different cardiovascular system components such as contractility, vascular impedance, atrio-ventricular synchronization, auto-regulation, etc, and perform 0D simulations of pressure and flow to investigate disease conditions, medical procedures, and physiologic states.

Our lab conducts research to construct, improve, and validate these physiology models. This can involve applying regression/machine learning to analyze clinical or literature data, performing optimization to tune the relevant model parameters, and/or implementing mathematical models to describe a specific physical scenario, medical procedure, or device operation.

A few examples of clinical-focused research projects involving the application of physiology modeling: 1) We introduced dysfunctions such as limited heart rate and impaired respiration into a virtual patient to study the resulting exercise performance; 2) We simulated steady versus pulsatile ventricular assist in patient-specific models to investigate why steady flow devices tends to produce better outcomes in neonates; 3) We modeled a range of patient conditions to determine the optimal surgical configuration and device setting for a cavopulmonary blood pump.

In-vitro Experiments & Testing

Materializing conceptual scenarios of interest into the real physical world is when rubber meets the road. The CMERL creates physiologically-realistic benchtop experimental environments to test the performance of medical devices and to predict the results of cardiovascular procedures. We design the test setups to mimic the cardiovascular system in various aspects depending on the relevant targets to be tested. Physical testing in these setups can provide important information towards the the design of devices and translation of novel treatment ideas before placing real patients at risk during clinical trials.

This research also pioneers new possibilities for personalized medicine, where in-vitro patient-specific testing can be done prior to medical decision, treatment, or device installation. Such a step forward could transform the current paradigms of clinical procedures.

Hybrid Experimental-Computational Modeling

NSF CAREER Award

Significant advances in biomedical science often leverage powerful computational and experimental modeling platforms. Our lab uses the hardware-in-the-loop approach to combine an in-vitro experiment with a numerical simulation of physiology, pioneering the Physiology Simulation COuPled Experiment (PSCOPE) which can capitalize on the strengths of both in-vitro and numerical platforms in a single hybrid model. This unified modeling framework enables a testing environment which simultaneously operates a medical device and performs computational simulations of the resulting physiology, providing a tool for physically testing medical devices and procedures with simulated physiologic feedback.

Our research in this area involves the continual advancement of the PSCOPE framework and the application of PSCOPE for clinical investigations. For example, we apply dynamic systems and control as well as optimization methods to achieve proper signal coupling between the computational and experimental domains in the PSCOPE model. We have also implemented the PSCOPE to model scenarios of Fontan cavopulmonary assist, acquiring physical measurements of assist device behaviour as well as simulated data of patient physiologic outcome.

Image-based Computational Fluid Dynamics

Each patient has unique blood flow characteristics based on their specific anatomy and physiology. The CMERL uses clinical imaging data to construct 3D models of patient-specific anatomy, as well as to determine flow conditions affecting the relevant anatomical regions. Coupling 3D models to systemic physiology, we perform multi-scale simulations to investigate the interactions between local and global hemodynamics. "Virtual surgeries" can also be performed computationally to predict the effects of different surgical scenarios and disease conditions. The applications of this research area include surgical design/prediction, risk stratification for adverse hemodynamic events, and biomechanics investigations.