CMBE2026 Invited Speakers

Computational Biomedical Engineering has great potential not only for innovation in medicine but, in my opinion, even for a true paradigm shift in health care. However, one must admit that clinically relevant success stories are still very limited. And this even though many groups with excellent people from all around the world have been working in this field for many years. Hence, I would like to use the honorable invitation to deliver the opening lecture of this edition of this fantastic conference series, not to report on our most recent work for modeling a certain organ or disease, but to take a broader perspective and will try to share some thoughts on how we might be better able to unleash the great potential that our field has to offer. In line with the core of this conference, I will share some recent modeling and methodical approaches that should be useful across a spectrum of questions in Biomedicine. But I will also discuss whether a shift in the general approach to addressing problems in Computational Biomedical Engineering can be useful. And I will even discuss some essential epistemological questions that I find crucial for the future of our field.

Wolfgang A. Wall is full Professor and founding Director of the Institute for Computational Mechanics at TU Munich. He received his PhD from the University of Stuttgart and acted as founding director of the Munich School of Engineering and is co-founder of two companies. He received several esteemed awards and serves on a large number of prestigious boards. Currently he also serves as Rector of CISM in Udine/Italy and is member of the Austrian and the Bavarian Academies of Sciences. His research can be described as “application motivated fundamental research” in a broad range of areas in computational mechanics, with a special focus on biophysics and biomedical engineering.

Congenital heart disease (CHD) is the most common birth defect and the leading cause of infant mortality in the US. CHD patients require a high degree of customization and individualized treatment planning, which is currently done “on the fly” in the operating room. Alternatively, personalized computational models now offer predictive decision support tools for designing bespoke treatments in advance of procedures. Recent advances in modeling and simulation are enabling increasingly sophisticated methods spanning 3D model construction, virtual surgery, and simulation-based outcomes prediction. State of the art methods include multi-physics cardiac models, virtual device implantation, vascular growth and remodeling, Bayesian approaches, and data-driven methods for image segmentation and digital twins. Clinical deployment of 3D modeling and simulation is now improving outcomes in CHD patients in a range of applications spanning valve reconstruction, biventricular repair, and interventional planning. Towards future developing of “living” implants, computational modeling is also supporting bioprinting applications, including design of vascular networks to optimize tissue perfusion at scale. The open source SimVascular project provides a software ecosystem for developing and sharing new capabilities with the broader scientific community.

Alison Marsden is the Douglass M. and Nola Leishman Professor of Cardiovascular Disease in the Departments of Pediatrics and Bioengineering at Stanford University. From 2007-2015 she was a faculty member at UCSD.  She graduated with a BSE degree from Princeton in 1998, and a PhD from Stanford Mechanical Engineering in 2005. She was a postdoctoral fellow at Stanford from 2005-07. She was the recipient of a Burroughs Wellcome Fund Career Award at the Scientific Interface in 2007, and an NSF CAREER award in 2011. She has been elected fellow of AIMBE, SIAM, APS, BMES, and ASME.  She is the 2023 recipient of the Van C. Mow medal from the ASME Bioengineering Division. She has published over 200 journal articles, and her research focuses on cardiovascular biomechanics and engineering methods to impact patient care in cardiac surgery and congenital heart disease.

Red blood cells (RBCs) tend to migrate toward the vessel axis in small vessels, forming the so-called axial accumulation. As a consequence, a region near the vessel wall that contains very few RBCs, known as the cell-free layer (CFL), is created. Smaller platelets subsequently migrate into this CFL, a phenomenon referred to as platelet margination. In this study, direct numerical simulations (DNS) were performed to investigate this phenomenon. We have previously conducted simulations of blood flow containing RBCs and platelet-like particles. In the present work, we employ the immersed boundary method to carry out DNS of blood flow with deformable RBCs and rigid platelet models. The effects of shear rate, channel size, and hematocrit on pressure-driven flow in straight and branched square channels are examined. The numerical results are compared with the experimental observations of Seki et al. (Sugihara-Seki, M., & Takinouchi, N. (2021). Margination of Platelet-Sized Particles in the Red Blood Cell Suspension Flow through Square Microchannels. Micromachines, 12(10), 1175). The DNS reveals that platelets, once transported to the corners of the square channel, remain there stably. This behavior is attributed to the small velocity fluctuations in the corner region. Based on the DNS data, we construct a model of the spatial distribution of velocity fluctuations. This distribution is incorporated as a random force term in a Langevin equation, where the amplitude of the random force varies spatially. Using this model, the characteristic corner-accumulation behavior of platelets is successfully reproduced.

Shu Takagi is a Professor at the Departments of Mechanical Engineering and Bioengineering at The University of Tokyo since 2010. He received Doctor of Engineering from the University of Tokyo in 1995. His areas of expertise include numerical simulations and experimental investigations on dispersed multiphase flows, especially bubbly flows and blood cell flows, medical ultrasound, hierarchical integrated simulation of human body, molecular thermo-fluid mechanics and multiscale analysis of thermo-fluid phenomena. He has written over 20 review articles including Annual Review of Fluid Mechanics and has given more than 30 keynote lectures in conferences. He has been contributing as an associate editor of IJMF and Journal FLOW. He is former presidents of Japan Society of Fluid Mechanics (2023-2024) and Japan Society of Multiphase Flows (2022-2023) and currently a member of Science Council of Japan. He is also a IUTAM Congress Committee member and contributing for the international activity in the field of mechanics.

Digital twins are adaptive virtual representations of physical systems that are updated by data and used to support prediction, monitoring, and decision-making. In biomedical applications, they may represent patients, organs, vascular systems, hospital workflows, or digital populations for treatment planning and virtual trials. Their reliability requires more than data and machine learning: it depends on physics-based models, scalable optimization algorithms, and uncertainty-aware decision support. This talk presents a PDE-constrained optimization perspective on reliable digital twins for biomedical and engineering systems. We begin with medical digital twins, focusing on patient-specific aneurysm modeling, where incomplete boundary conditions, uncertain inflow profiles, and truncated geometries can strongly affect clinically relevant flow quantities. Adjoint-based sensitivity analysis provides a principled way to quantify these effects and guide model improvement. The same framework extends to structural health monitoring, thermal design, neuromorphic sensing, graph-based control, and Maxwell-based digital twins. Across these examples, PDE-constrained optimization provides a common language for state estimation, parameter identification, control, sensor placement, and model updating. The central message is that trustworthy digital twins are not merely data-driven models; they are optimization-driven, uncertainty-aware, structure-preserving computational systems that close the loop between measurements, prediction, and decisions.

Harbir Antil is Professor of Mathematics and Director/Founder of the Center for Mathematics and Artificial Intelligence (CMAI) at George Mason University. His research lies at the intersection of optimization, partial differential equations (PDEs), and scientific computing, with a focus on PDE-constrained and nonsmooth optimization, uncertainty quantification, and reduced-order and hybrid physics–AI modeling. He has pioneered new algorithmic and analytical frameworks for risk-averse and nonlocal PDE optimization, and his recent work has advanced the mathematical foundations of digital twins, linking rigorous PDE theory with data assimilation, control, and decision making. Dr. Antil’s research is broadly supported by the NSF, DOE, DARPA, ONR, AFOSR, DTRA, and industry partners including Siemens and Simula Labs. His publications appear in leading SIAM journals, and his work has been featured in the New York Times and NSF Research Highlights. He is a member of Intelligence Science and Technology Experts Group at National Academies of Sciences. He serves as Associate Editor for the SIAM Journal on Scientific Computing and on the SIAM Review Section, and is President of the SIAM DC–MD–VA Section. Beyond academia, his optimization and digital twin frameworks have influenced applications in structural health monitoring, biomedical systems, and neuromorphic imaging, with algorithms deployed aboard the International Space Station. He is an invited plenary speaker at the Joint Mathematics Meetings 2026, reflecting his international stature in applied and computational mathematics.

Cardiovascular diseases remain the leading cause of morbidity and mortality worldwide, underscoring the need for more personalized strategies in prevention, diagnosis, and treatment. Conventional cardiovascular risk prediction models are based on population-level statistics and limited clinical variables, often neglecting individual physiological diversity. Recent advances in artificial intelligence (AI), medical image analysis, and in-silico modeling are transforming how cardiovascular diseases are studied and managed. AI-techniques (e.g., radiomics, deep learning) and digital twin frameworks enable the extraction of high-dimensional features from multimodal data, supporting non-invasive, individualized assessment of cardiovascular structure, function, and disease progression. This talk will present applications of these computational and data-driven approaches in several translational cardiovascular research areas. Examples include in-silico clinical trials for acute ischemic stroke; multi-organ digital twins to explore mechanistic links between cardiac function and cerebrovascular outcomes; AI-driven, personalized prevention strategies addressing sex-specific cardiovascular risk; and multimodal modelling integrating heterogeneous data modalities to generate personalized risk profiles and decision-support predictions. Collectively, these applications illustrate how integrative digital technologies combining imaging, simulation, and clinical data can enhance disease understanding, improve predictive accuracy, and advance precision cardiovascular medicine. The discussion will also highlight international initiatives addressing these challenges.

Dr. Nerea Arrarte Terreros is currently an Associate Researcher at the Digital Health and Biomedical Technologies department of Vicomtech, a Spanish Research and Technology Organization specialized in Artificial Intelligence and Computer Vision. She is also an affiliated supervisor in the Cardiology and Biomedical Engineering & Physics departments of the Amsterdam University Medical Centers (Amsterdam UMC), and co-founder of inSteps B.V., a spin-off from Amsterdam UMC developing novel in-silico technologies to support medical device development for stroke care. With a background in physics, her research integrates medical image analysis, artificial intelligence, and computational modeling to advance personalized and translational approaches in healthcare. She has a strong track record in European-funded initiatives. Her research has been recognized with multiple international mobility and research excellence awards. She is also actively involved in innovation valorization and collaborative projects with industry partners to accelerate the translation of computational technologies into clinical practice.

Cardiovascular computational biomechanics, particularly those employing patient- specific information such as CT and/or MRI images of patients, has evolved dramatically in last 2 decades, aimed at contributing to clinical practice. Continuous efforts to improve physiological fidelity and computational efficiency have led to the development of several diagnostic and prognostic systems based on computational biomechanics, several of which are now commercially available and integrated into routine clinical workflows. This talk will showcase our recent effort towards translating cardiovascular computational biomechanics to underpin clinical science and practice. Special focus will be placed on three key areas: (1) streamlining computational workflows and acceleration by reduced order models and machine learning; (2) insights from large patient cohort studies; and (3) model parameter inference for effective patient digital twinning. The first two topics will be illustrated through studies on coronary artery atherosclerotic disease and intervention with stents, whilst the third will focus on patients with heart failure supported by left ventricular assist devices. The talk will also address unique challenges involved in these translational efforts, such as to identify the optimal balance between analysis speed, physiological representation and practicality

Professor Ryo Torii is Professor of Biomechanics in the Department of Mechanical Engineering at University College London, where he leads the Biomechanical Engineering Research Division. His research focuses on the interplay between mechanics and physiology, with particular emphasis on the transition of the human cardiovascular system from health to disease, including post-intervention states. With over 170 peer- reviewed publications and book chapters, he has made marked contribution to cardiovascular biomechanics and clinical science, especially in modelling the coronary circulation in relation to atherosclerotic diseases. His long-term research ambition is to strengthen the scientific foundations of clinical practice, which has been pursued through the development of novel computational approaches and extensive international collaboration, involving 37 institutions in 18 countries. His most recent work continues to advance clinical translation via the development of virtual patient simulators with PAC Dynamic Inc., a start-up company he co-founded, and of an airflow optimisation tool for NHS Hospitals in the UK (NIHR Air Safety Programme).

Dynamic imaging has become a cornerstone of functional assessment in bioengineering, providing time-resolved information on perfusion, flow, and tracer transport across organs and tissues. However, the interpretation of these data is often limited by indirect observability, noise, and the complexity of underlying physiological processes. Computational flow and transport modeling offers a powerful framework to overcome these limitations, enabling mechanistic interpretation of dynamic imaging signals and supporting patient-specific prediction.

Recent advances demonstrate a fruitful cross-talk between dynamic imaging and modeling, in which imaging informs model personalization, while models extend imaging from descriptive analysis toward quantitative, predictive biomarkers. Compartmental and reduced-order models derived from first principles provide a rigorous basis for tracer-kinetic analysis and perfusion estimation, with important implications for model identifiability and validation in dynamic imaging contexts. When integrated into multiscale pipelines, these models can be coupled with three-dimensional flow simulations and zero-dimensional cardiovascular representations, enabling digital twin workflows driven by patient-specific data.

Such approaches facilitate sensitivity analysis, parameter estimation, and uncertainty quantification, thereby improving robustness and clinical interpretability of computational predictions. Applications range from evaluation of liver graft perfusion and viability using fluorescence imaging, to tumor microcirculation assessment via contrast-enhanced imaging, highlighting the translational potential of imaging-driven modeling frameworks. By unifying dynamic imaging with flow and transport modeling, this cross-disciplinary synergy advances predictive bioengineering toward more reliable, patient-specific, and clinically actionable tools.

Irene Vignon-Clementel is a Research Director (Full Professor equiv.) at Inria, the French National Institute for Research in Digital Science and Technology. Her research group focuses on developing virtual human twins of physiological flows and transport. The aim is to better understand certain pathologies and improve their treatment (procedure/surgical planning, medicaldevice design). This is enabled by mechanistic and learning mathematicalmodelling and medical imaging. IVC has been awarded an ERC consolidator grant and is scientific coordinator of the Horizon Europe project ARTEMIs. She has published over 100 articles and book chapters, and is co-inventor of 2 patents. She has been a member of various scientific councils – 3DS-FDA ENRICHMENT Project (USA), the BIOREME network (UK), PariSantéCampus (France) – and editorial boards (International Journal onNumerical Methods in Biomedical Engineering and Royal Society OpenScience). She sits on the board of directors of the Virtual PhysiologicalHuman Institute. She is actively promoting the modeling and medicine interface through co-supervision of MD-PhDs, joint research projects, conference organization and interface articles with clinicians.

Magnetic resonance imaging (MRI) is a well-established tool for noninvasively acquiring internal body information and has been extended to flow velocity imaging such as phase- contrast imaging (PCI) and four-dimensional flow MRI. More recently, diffusion-weighted imaging (DWI) and derived metrics such as the apparent diffusion coefficient (ADC) have been applied to probe slow cerebrospinal fluid (CSF) transport in the intracranial region. However, MRI-based flow measurements are affected by artifacts that depend on the underlying flow profiles and MRI scan settings, and the relationships between different MRI flow-encoding techniques (e.g., PCI and DWI) remain poorly understood. This lack of a unified physical interpretation makes accurate in vivo flow quantification challenging. To address this issue, we are developing computational modeling frameworks for MRI flow measurements that integrate multiple MRI flow-encoding techniques. Our recent studies have provided quantitative estimates of imaging artifacts as functions of the target flow profiles and scan conditions and have begun to connect PCI and DWI from a fluid- mechanical perspective. This framework offers a unified basis for interpreting MRI flow measurements and has the potential to expand quantitative MRI-based flow assessment toward future clinical applications.

Tomohiro Otani is an Associate Professor in the Department of Mechanical Science and Bioengineering, Graduate School of Engineering Science, The University of Osaka, Japan. His research focuses on computational biomechanics with a strong emphasis on clinical translation, particularly the modeling of intracranial neurovascular blood flow and cerebrospinal fluid dynamics. His recent work aims to unify MRI physics and continuum mechanics to enable quantitative in vivo flow evaluation and digital twins of neurofluid systems. He also holds a cross-appointment as a Guest Associate Professor in the Department of Radiology at Shiga University of Medical Science, promoting interdisciplinary collaborations between mechanical engineering and clinical radiology. Over the decade following his PhD, he has published around 50 peer-reviewed papers and book chapters, and has received several recognitions, including the Young Scientist Award in Science and Technology from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, and the Yamaguchi Medal from the Asian-Pacific Association for Biomechanics.

The treatment of cardiovascular disease places a substantial burden on global healthcare systems. Vascular pathologies such as atherosclerosis, aortic dissections, and aneurysms are all linked to rupture of vessel wall tissue, often with fatal consequences. Predicting, and ultimately preventing, such ruptures require further understanding of vessel wall fracture mechanisms. Given its hierarchical structure, vascular tissue exhibits complex mechanical properties, therefore necessitating robust and efficient numerical tools to describe its fracture mechanics. This presentation summarizes efforts in characterizing aortic wall fracture, where especially the vessel wall’s anisotropic mechanical properties are addressed. A number of visco-hyperelastic finite-deformation models described vessel wall tissue and numerical concepts, such as cohesive zone modeling, Extended Finite Element Method (XFEM) modeling and phase-field approaches have been explored. Our previously established in-vitro symmetry-constrained Compact Tension (symconCT) test combined with Digital Image Correlation (DIC) provided rich information for model development, calibration and validation. Loading porcine aorta specimens in the axial direction resulted in mode I tissue failure and a fracture path aligned with the circumferential vessel direction. While circumferential loading still resulted in mode I-dominated failure, the fracture path displayed multiple deflections. In highly diseased specimens from the human aneurysmatic aorta, fracture is often diverted along the circumferential direction, an experimental observation that could only be captured by an anisotropic phase field description. Regardless phase field modelling allows to deal with most complexity of vascular tissue fracture, it requires significant numerical resources. Current numerical approaches would therefore need considerable further development to assess the risk profile of vascular pathologies towards patient-specific health care decision making.

T.Christian Gasser is Professor of Biomechanics at KTH Royal Institute of Technology, Stockholm, where leads the soft biological tissue research group. He holds a Master of Mechanical Engineering (1997) and a PhD in Civil Engineering (2001), both from Graz University of Technology, Austria. His main research interest concerns the development and application of advanced numerical techniques to solve realistic (bio)engineering and clinical problems. Constitutive models for anisotropic finite strain materials, developed as part of his PhD thesis, have been implemented in all major Finite Element Method (FEM) simulation packages, and Gasser’s translational biomechanics research led to A4clinicsRE, commercial biomechanical-based simulation software for clinical decision making.  He is listed as KTH’s most influential researcher in Biomedical Engineering (more than 20k Google Scholar citations with an h-index of 59), in 2022 he was awarded a Humboldt Research Award from Germany, and 2024 he delivered the Odqvist lecture, a distinction awarded by the Swedish national mechanics committee. He is principal founder of ARTEC Diagnosis AB as well as VASCOPS GmbH and serves a legal expert for skiing accident reconstruction at Oberlandesgericht, Graz, Austria.