• Efficient multi-fidelity fluid-structure interaction modeling for pulsatile blood flow in deformable biological tissues (C.M. Lee et al. , Annals of Biomedical Engineering, 2026)

Simulating tissue deformation and flow alterations under external compression typically requires computationally expensive fluid–structure interaction (FSI) analysis. We present a multi-fidelity FSI framework that efficiently captures tissue mechanics and hemodynamic responses to dynamic pressure and demonstrate its application to intermittent pneumatic compression (IPC). The model couples a one-dimensional deformable blood flow formulation with the three-dimensional Cauchy equation of motion. Compared with full 3D FSI, the framework achieved <1% flow-rate error in both simplified and subject-specific geometries and <2% pressure error in the simplified case, while reducing computational cost by factors of 9 and 46, respectively. In IPC simulations, a full pulsatile cycle was completed in 457 seconds (simplified) and 42.2 minutes (subject-specific). This approach enables accurate, computationally efficient simulation under external compression, supporting large-scale parametric studies and therapeutic device optimization. 

• Multiscale Transport Modeling in Biological Tissue (H. Song et al., to be submitted to Computer Methods in Applied Mechanics and Engineering)

Heat and mass transport in biological tissue are strongly influenced by blood flow, vascular architecture, and tissue heterogeneity. However, conventional bioheat and mass transport models often treat tissue as a homogeneous or porous medium, limiting their ability to capture localized vessel–tissue interactions. We propose a multiscale transport framework that explicitly models larger vessels using one-dimensional flow and transport equations, while representing smaller vessels implicitly within the tissue continuum. The one-dimensional vascular and three-dimensional tissue domains are explicitly coupled to ensure bidirectional exchange and conservation of energy or mass. The framework is validated against analytical solutions and commercial solver results and compared with conventional bioheat models to assess differences between explicit and homogenized vascular representations. Applications to localized regions with complex vascular networks demonstrate its capability under physiologically relevant conditions. 

• Reduced-order model for pulsatile flow simulations (W. Choi et al., Computer Methods and Programs in Biomedicine, 2025)

This study presents a one-dimensional reduced-order model (ROM) for pulsatile blood flow in which the pressure–flow relationship is calibrated directly from a minimal number of three-dimensional CFD simulations for each vascular geometry. The model preserves geometry-dependent hemodynamics while achieving significant computational speedup. Validation against full 3D simulations shows that the ROM accurately reproduces pressure and flow waveforms in both idealized and subject-specific vessels.

• Uncertainty quantification of the Fractional Flow Reserve (W. Choi et al., under review in the Annals of Biomedical Engineering)

Fractional flow reserve (FFR) is a functional index used to assess the presence and severity of myocardial ischemia in coronary arteries. This study applies the calibrated 1D coronary ROM to quantify uncertainty in FFR under realistic pulsatile physiology. Global sensitivity analysis identifies intramyocardial pressure effects, myocardial contractility, and systolic duration as dominant factors, while results demonstrate that FFR is generally robust to physiological uncertainty, with variability increasing systematically as stenosis severity increases and FFR decreases.

• Multiscale modeling of posture-dependent cerebrovascular hemodynamics with autoregulatory coupling (H.J. Kim et al., Computers in Biology and Medicine, 2026)

Cerebral blood flow is maintained by autoregulatory mechanisms that compensate for systemic and postural changes. We investigated the coupled effects of aortic pressure, body posture, and vessel wall stiffness on cerebrovascular hemodynamics using a multiscale model incorporating arterial and venous networks down to the pre- and postcapillary levels, coupled with a three-dimensional perfusion domain. The framework integrates passive vessel mechanics and active autoregulation to simulate pressure- and metabolism-driven arteriolar responses. Simulations across aortic pressures (30–150 mmHg), postures (supine, upright, inverted), and varying wall stiffness showed that arterial deformation and total vascular volume were sensitive to pressure and gravity, while venous volume was primarily posture-dependent. Autoregulation maintained near-constant cerebral blood flow by dynamically adjusting arteriolar diameter. Increased compliance amplified posture-induced volume changes and autoregulatory responses, whereas higher stiffness attenuated both. This framework clarifies the interplay between vascular mechanics and autoregulation in stabilizing cerebral perfusion under varying physiological conditions.

• A Stress-to-Strength Metric for Rupture Risk Assessment in Abdominal Aortic Aneurysms (C. An et al., to be submitted to Computers in Biology and Medicine)

Abdominal aortic aneurysm (AAA) rupture remains a major cause of morbidity and mortality, yet current clinical indices inadequately capture the complex biomechanical mechanisms underlying rupture. We propose a dimensionless stress-to-strength ratio (SSR) that integrates pulsatile loading and material fatigue into a unified metric, enabling consistent assessment of rupture susceptibility across heterogeneous wall conditions. Twenty-three idealized AAA models with varied geometric, material, and physiological parameters were analyzed using three-dimensional one-way fluid–structure interaction simulations. SSR distributions were compared with conventional stress- and strain-based metrics and evaluated against clinically reported rupture patterns. SSR consistently identified all rupture-prone regions, whereas conventional metrics captured only a subset. Elevated rupture risk was associated with thick intraluminal thrombus (ILT), thin ILT-free segments, sagging ILT regions, increased blood pressure, reduced wall thickness, and tissue softening. Overall, SSR more effectively captures the interplay between mechanical loading and tissue strength, improving prediction of AAA rupture patterns and demonstrating strong potential as a clinically relevant biomechanical indicator.

• Computational investigation of intermittent pneumatic compression operating parameters and tissue mechanics on lower-limb venous hemodynamics (Y.J. Choi et al., to be submitted to Computers in Biology and Medicine)

Intermittent pneumatic compression (IPC) is widely used to manage chronic venous insufficiency, yet the relationships among operating parameters, tissue mechanics, and venous hemodynamics remain unclear. We employed a multi-fidelity fluid–structure interaction framework coupling one-dimensional deformable blood flow with three-dimensional tissue mechanics and a lumped upstream venous network to efficiently evaluate these effects. Parametric analyses were performed in simplified and subject-specific geometries, varying chamber number, hold time, maximum compression pressure, and tissue properties. Maximum compression pressure and hold duration were the primary determinants of venous flow, pressure, and wall shear stress (WSS). Higher pressures increased lumen narrowing and flow variation, while longer hold times promoted distal blood accumulation and elevated peak flow and pressure. Chamber number had a minor influence under uniform actuation. More compliant tissues exhibited greater deformation and stronger hemodynamic responses. IPC consistently increased WSS due to simultaneous flow augmentation and radius reduction. These results provide quantitative insight into IPC-induced venous modulation and highlight the importance of patient-specific parameter optimization for personalized compression therapy.

• Computational Modeling of Collateral Circulation and Arteriolar Vasodilation in Ischemic Tissue Perfusion (N. Moreaux et al., Annals of Biomedical Engineering, 2026)

To better capture physiological responses under pathological conditions, we incorporate two key compensatory mechanisms—collateral circulation and arteriolar vasodilation—into a multiscale blood perfusion framework. Collateral vessels are stochastically generated to supply ischemic regions, with density, caliber, and distribution systematically varied. Arteriolar vasodilation is modeled to represent vascular smooth muscle responses to metabolic and hemodynamic stimuli. Artificial stenoses are introduced in both idealized and subject-specific cerebrovascular geometries to assess perfusion changes. Collateral vessels alone provide limited perfusion improvement, strongly dependent on vessel caliber and density. In contrast, combining collateral flow with vasodilation significantly enhances tissue perfusion, indicating that vasodilation is the dominant compensatory mechanism, with collaterals providing secondary support. The proposed framework enables efficient multiscale simulations that incorporate major autoregulatory mechanisms and yields results consistent with literature data.

• Computational Analysis of Cerebral Hemodynamics Under Variations in Autoregulation and Circle of Willis Anatomy (S.Y. Kim et al., to be submitted to Computers in Biology and Medicine)

Patient-specific cerebral hemodynamics were modeled using the incompressible Navier–Stokes equations. Clinical imaging data were incorporated to construct boundary condition subroutines that capture cerebrovascular autoregulatory mechanisms, integrating key physiological factors governing vessel diameter regulation. This framework enables accurate prediction of individual hemodynamic responses and supports quantitative assessment of disease progression risk.

• Computational Methods and Applications for Diagnosing Chronic Total Occlusion (W. Choi et al., to be submitted to Journal of the American College of Cardiology) 

• Investigation of More Sustainable Nuclear Energy System Using Spent Fuel Chlorination & Molten Salt Fast Reactors

• Development and Validation of Cardiovascular and Cerebrovascular M3DT Using Patient-Specific Diagnostic Technology, Ministry of Food and Drug Safety  (개인 맞춤형 의료진단기술 기반의 심뇌혈관 M3DT 기술개발 및 검증, 보건산업진흥원, 주관)

• Multiscale Brain Mechanics Model for Digital Twin Technology-based Monitoring of the Brain, Samsung Research Foundation (디지털 트윈 기술 기반 뇌 모니터링을 위한 멀티 스케일 뇌 역학 모델, 삼성미래기술육성사업, 주관)

• Research Center for Precision Medicine Platform Based on Smart Hemo-Dynamic Index, National Research Foundation (스마트 혈류역학 지표 기반 실시간 정밀의료 플랫폼 연구센터, 한국연구재단, 공동)

• CNS focused Fluid Shift Driven Diseases, Ministry of Health and Wellfare (우주환경내 체액이동에 의한 중추신경계 질환규명 및 진단치료기술, 보건복지부, 공동)

• Investigation of More Sustainable Nuclear Energy System Using Spent Fuel Chlorination & Molten Salt Fast Reactors, KAIST (사용후연료 염소화 및 용융염고속로를 이용한 보다 지속가능한 원자력 시스템 개념 연구, 한국과학기술원, 공동)