MS03 - CARD-02

Novel multiscale and multisystem approaches to cardiovascular modeling and simulation (Part 3)

Tuesday, July 15 at 10:20am

SMB2025 Follow

Note: this minisymposia has multiple sessions. The other sessions are: Part-1, Part-2, and Part-4.

Mitchel J. Colebank (University of South Carolina), Vijay Rajagopal, The University of Melbourne, Australia

Description:

Cardiovascular models are now recognized as a potential frontier in the development of personalized models and digital twins. This new excitement in the field is fueling new strides in mathematical and computational approaches to describe cardiovascular function across different temporal and/or spatial scales. In addition, new multisystem models accounting for how the heart and vasculature interact with other organ systems are being developed in combination with tools located at the heart of data science. Thus, this minisymposium will focus on the development of cardiovascular models that mimic cardiovascular function across spatial or temporal scale, new models that couple the cardiovascular system with other physiological systems, and new innovations in data-driven solutions to modeling cardiovascular phenomena. Potential topics include: - Multiscale models of cardiac and vascular function; - Computational approaches to cell-tissue-organ level function; - Mathematical coupling of cardiac and vascular mechanics; - Modeling autonomic control and neurovascular function; - Simulating tissue growth and remodeling; - Multisystems models of cardiovascular-organ interactions; and - Physics-informed data science approaches to cardiovascular science

Mette Olufsen

North Carolina State University

"An uncertainty aware framework for generating vascular networks from imaging"

A well-calibrated mathematical model and an understanding of uncertainties in model predictions are essential for generating a digital twin. Creating a patient-specific cardiovascular model typically involves two key steps: (i) constructing the vascular domain and (ii) performing hemodynamic simulations. The vascular domain is usually obtained by segmenting CT or MRI scans to reconstruct the vascular network. Once constructed, hemodynamic simulations are conducted using inferred model parameters that minimize discrepancies between computed results and available physiological data. This talk will addres challenges in generating 1D network models with multiple branching generations and detecting abnormalities within these networks. One significant challenge is the automatic extraction of vessel centerlines, which is crucial for 1D modeling. We focus on a skeletonization-based method for centerline extraction, which iteratively removes voxels until only a single-voxel-wide path remains within each vessel. Using statistical change-point analysis, we construct a labeled directed graph (a tree) that encodes vessel connectivity, radii, and lengths. By sampling from normal distributions of these quantities with a 1D fluid dynamics model, we explore how uncertainties in geometry affect hemodynamic predictions. Our results emphasize the importance of accounting for image-based uncertainty in medical modeling.

Sara Johnson

University of Puget Sound

"Modeling Microglial Response to MCAO-Induced Ischemic Stroke"

Neuroinflammation immediately follows the onset of ischemic stroke in the middle cerebral artery. During this process, microglial cells are activated in and recruited to the penumbra. Microglial cells can be activated into two different phenotypes: M1, which can worsen brain injury; or M2, which can aid in long-term recovery. In this study, we contribute a summary of experimental data on microglial cell counts in the penumbra following ischemic stroke induced by middle cerebral artery occlusion (MCAO) in mice and compile available data sets into a single set suitable for time series analysis. Further, we formulate a mathematical model of microglial cells in the penumbra during ischemic stroke due to MCAO. Through use of global sensitivity analysis and Markov Chain Monte Carlo (MCMC)-based parameter estimation, we analyze the effects of the model parameters on the number of M1 and M2 cells in the penumbra and fit identifiable parameters to the compiled experimental data set. We utilize results from MCMC parameter estimation to ascertain uncertainty bounds and forward predictions for the number of M1 and M2 microglial cells over time. Results demonstrate the significance of parameters related to M1 and M2 activation on the number of M1 and M2 microglial cells. Simulations further suggest that potential outliers in the observed data may be omitted and forecast predictions suggest a lingering inflammatory response.

Simon Walker-Samuel

University College London

"Using physics-informed deep generative learning to model blood flow in the retina"

Disruption of retinal vasculature is linked to various diseases, including diabetic retinopathy and macular degeneration, leading to vision loss. We present here a novel algorithmic approach that generates highly realistic digital models of human retinal blood vessels, based on established biophysical principles, including fully-connected arterial and venous trees with a single inlet and outlet. This approach, using physics-informed generative adversarial networks (PI-GAN), enables the segmentation and reconstruction of blood vessel networks with no human input and which out-performs human labelling. Segmentation of DRIVE and STARE retina photograph datasets provided near state-of-the-art vessel segmentation, with training on only a small (n = 100) simulated dataset. Our findings highlight the potential of PI-GAN for accurate retinal vasculature characterization, with implications for improving early disease detection, monitoring disease progression, and improving patient care.

Mitchel Colebank

University of South Carolina

"Effects of vasomotor tone on systemic vascular wave reflections"

One-dimensional, pulse-wave propagation models are able to replicate hemodynamic waveforms that are representative of measured data. While these models are a potential tool in the era of digital twins, few models have considered the role of smooth muscle vasoactivity and its effects on blood pressure and flow. This is especially important for understanding cerebrovascular function, especially in diseases like dementia and Alzheimer's, where cerebral vasoactivity is known to be a cause and consequence of altered mechanical stimuli. Thus, there is a need for new computational models that explicitly account for vascular tone during hemodynamic simulation. Here, we implement a relatively simplistic exponential model of the proximal vasculature pressure-area relationship which incorporates extracellular matrix stiffness, vascular smooth muscle cell stiffness, the degree of vasomotor tone in comparison to some reference tone, and the reference pressure. We couple this vasoactive large vessel model to the structured tree boundary condition, which represents the microvascular beds. To differentiate between proximal and small vessel vasoconstriction, we also introduce a vasodilation factor in the structured tree that controls microvascular radii. We analyze the model using global sensitivity analysis, and provide insight into the distinct contributions of large and small vessel vasoactivity in an idealized systemic arterial network. Our results show that microvascular vasoconstriction is more impactful that proximal vessel vasotone, but that stress-strain behavior in the large vessels can be modulated divergently depending on the relative magnitudes of ECM and smooth muscle stiffness. This study lays the foundation for future studies investigating the effects of vasoactivity on hemodynamic outcomes.

#SMB2025 Follow

Annual Meeting for the Society for Mathematical Biology, 2025.