Computational hemodynamics
At WIAS, we improve methods in computational hemodynamics and develop methods for optimal design and fluidstructure interaction problems in order to enhance the understanding of blood physiology and pathological circulation dynamics,
 allow efficient patientspecific blood flow simulations, e.g. for therapy planning,
 estimate vessel parameters by data assimilation,
 optimally design cardiovascular implants.
Data assimilation in onedimensional networks
Simulated blood flow in the aorta
Mathematical modelling of tissues
The blood flow dynamics in the human heart depends highly on the deformation of the surrounding materials as for instance the heart muscle, elastic vessels, the pericardium etc. In connection with fluidstructure interaction as well as multiscale modelling and cancer research, we investigate effects of calcium release and transport in cardiac muscles,
 noninvasive estimation techniques of elastic tissue properties,
 modelling, analysis and simulation of contact phenomena in the pericardium,
 tumour growth and vascularization.
Publications
Monographs

A. Caiazzo, I.E. VignonClementel, Chapter 3: Mathematical Modeling of Blood Flow in the Cardiovascular System, in: Quantification of Biophysical Parameters in Medical Imaging, I. Sack, T. Schaeffter, eds., Springer International Publishing, Cham, 2018, pp. 4570, (Chapter Published), DOI 10.1007/9783319659244_3 .
Articles in Refereed Journals

G. Dong, M. Hintermüller, K. Papafitsoros, Quantitative magnetic resonance imaging: from fingerprinting to integrated physicsbased models, SIAM Journal on Imaging Sciences, 2 (2019), pp. 927971, DOI 10.1137/18M1222211 .
Abstract
Quantitative magnetic resonance imaging (qMRI) is concerned with estimating (in physical units) values of magnetic and tissue parameters, e.g., relaxation times $T_1$, $T_2$, or proton density $rho$. Recently, in [Ma et al., Nature, 495 (2013), pp. 187193], magnetic resonance fingerprinting (MRF) was introduced as a technique being capable of simultaneously recovering such quantitative parameters by using a twostep procedure: (i) given a probe, a series of magnetization maps are computed and then (ii) matched to (quantitative) parameters with the help of a precomputed dictionary which is related to the Bloch manifold. In this paper, we first put MRF and its variants into perspective with optimization and inverse problems to gain mathematical insights concerning identifiability of parameters under noise and interpretation in terms of optimizers. Motivated by the fact that the Bloch manifold is nonconvex and that the accuracy of the MRFtype algorithms is limited by the ?discretization size? of the dictionary, a novel physicsbased method for qMRI is proposed. In contrast to the conventional twostep method, our model is dictionaryfree and is rather governed by a single nonlinear equation, which is studied analytically. This nonlinear equation is efficiently solved via robustified Newtontype methods. The effectiveness of the new method for noisy and undersampled data is shown both analytically and via extensive numerical examples, for which improvement over MRF and its variants is also documented. 
J. Sprekels, H. Wu, Optimal distributed control of a CahnHilliardDarcy system with mass sources, Applied Mathematics and Optimization. An International Journal with Applications to Stochastics, published online on 24.01.2019, DOI 10.1007/s00245019095554 .
Abstract
In this paper, we study an optimal control problem for a twodimensional CahnHilliardDarcy system with mass sources that arises in the modeling of tumor growth. The aim is to monitor the tumor fraction in a finite time interval in such a way that both the tumor fraction, measured in terms of a tracking type cost functional, is kept under control and minimal harm is inflicted to the patient by administering the control, which could either be a drug or nutrition. We first prove that the optimal control problem admits a solution. Then we show that the controltostate operator is Fréchet differentiable between suitable Banach spaces and derive the firstorder necessary optimality conditions in terms of the adjoint variables and the usual variational inequality. 
S.P. Frigeri, C.G. Gal, M. Grasselli, J. Sprekels, Strong solutions to nonlocal 2D CahnHilliardNavierStokes systems with nonconstant viscosity, degenerate mobility and singular potential, Nonlinearity, 32 (2019), pp. 678727, DOI 10.1088/13616544/aaedd0 .
Abstract
We consider a nonlinear system which consists of the incompressible NavierStokes equations coupled with a convective nonlocal CahnHilliard equation. This is a diffuse interface model which describes the motion of an incompressible isothermal mixture of two (partially) immiscible fluids having the same density. We suppose that the viscosity depends smoothly on the order parameter as well as the mobility. Moreover, we assume that the mobility is degenerate at the pure phases and that the potential is singular (e.g. of logarithmic type). This system is endowed with noslip boundary condition for the (average) velocity and homogeneous Neumann boundary condition for the chemical potential. Thus the total mass is conserved. In the twodimensional case, this problem was already analyzed in some joint papers of the first three authors. However, in the present general case, only the existence of a global weak solution, the (conditional) weakstrong uniqueness and the existence of the global attractor were proven. Here we are able to establish the existence of a (unique) strong solution through an approximation procedure based on time discretization. As a consequence, we can prove suitable uniform estimates which allow us to show some smoothness of the global attractor. Finally, we discuss the existence of strong solutions for the convective nonlocal CahnHilliard equation, with a given velocity field, in the three dimensional case as well. 
L. Blank, A. Caiazzo, F. Chouly, A. Lozinski, J. Mura, Analysis of a stabilized penaltyfree Nitsche method for the Brinkman, Stokes, and Darcy problems, ESAIM: Mathematical Modelling and Numerical Analysis, 52 (2018), pp. 21492185, DOI 10.1051/m2an/2018063 .

A. Caiazzo, F. Caforio, G. Montecinos, L.O. Müller, P.J. Blanco, E.F. Toro, Assessment of reduced order Kalman filter for parameter identification in onedimensional blood flow models using experimental data, International Journal of Numerical Methods in Biomedical Engineering, 33 (2017), pp. e2843/1e2843/26, DOI 10.1002/cnm.2843 .
Abstract
This work presents a detailed investigation of a parameter estimation approach based on the reduced order unscented Kalman filter (ROUKF) in the context of onedimensional blood flow models. In particular, the main aims of this study are (i) to investigate the effect of using real measurements vs. synthetic data (i.e., numerical results of the same in silico model, perturbed with white noise) for the estimation and (ii) to identify potential difficulties and limitations of the approach in clinically realistic applications in order to assess the applicability of the filter to such setups. For these purposes, our numerical study is based on the in vitro model of the arterial network described by [Alastruey et al. 2011, J. Biomech. bf 44], for which experimental flow and pressure measurements are available at few selected locations. In order to mimic clinically relevant situations, we focus on the estimation of terminal resistances and arterial wall parameters related to vessel mechanics (Young's modulus and thickness) using few experimental observations (at most a single pressure or flow measurement per vessel). In all cases, we first perform a theoretical identifiability analysis based on the generalized sensitivity function, comparing then the results obtained with the ROUKF, using either synthetic or experimental data, to results obtained using reference parameters and to available measurements. 
C. Bertoglio, A. Caiazzo, A Stokesresidual backflow stabilization method applied to physiological flows, Journal of Computational Physics, 313 (2016), pp. 260278.
Abstract
In computational fluid dynamics incoming flow at open boundaries, or emphbackflow, often yields to unphysical instabilities for high Reynolds numbers. It is widely accepted that this is due to the incoming energy arising from the convection term, which cannot be empha priori controlled when the velocity field is unknown at the boundary. In order to improve the robustness of the numerical simulations, we propose a stabilized formulation based on a penalization of the residual of a weak Stokes problem on the open boundary, whose viscous part controls the incoming convective energy, while the inertial term contributes to the kinetic energy. We also present different strategies for the approximation of the boundary pressure gradient, which is needed for defining the stabilization term. The method has the advantage that it does not require neither artificial modifications or extensions of the computational domain. Moreover, it is consistent with the Womersley solution. We illustrate our approach on numerical examples  both academic and reallife  relevant to blood and respiratory flows. The results also show that the stabilization parameter can be reduced with the mesh size. 
A. Caiazzo, R. Guibert, I.E. VignonClementel, A reducedorder modeling for efficient design study of artificial valve in enlarged ventricular outflow tracts, Computer Methods in Biomechanics and Biomedical Engineering, 19 (2016), pp. 13141318.
Abstract
A computational approach is proposed for efficient design study of a reducer stent to be percutaneously implanted in enlarged right ventricular outflow tracts (RVOT). The need for such a device is driven by the absence of bovine or artificial valves which could be implanted in these RVOT to replace the absent or incompetent native valve, as is often the case over time after Tetralogy of Fallot repair. Hemodynamics are simulated in the stented RVOT via a reduce order model based on proper orthogonal decomposition (POD), while the artificial valve is modeled as a thin resistive surface. The reduced order model is obtained from the numerical solution on a reference device configuration, then varying the geometrical parameters (diameter) for design purposes. To validate the approach, forces exerted on the valve and on the reducer are monitored, varying with geometrical parameters, and compared with the results of full CFD simulations. Such an approach could also be useful for uncertainty quantification. 
A. Caiazzo, R. Guibert, Y. Boudjemline, I.E. VignonClementel, Efficient blood flow simulations for the design of stented valve reducer in enlarged ventricular outflow tracts, Cardiovascular Engineering and Technology, 6 (2015), pp. 485500.
Abstract
Tetralogy of Fallot is a congenital heart disease characterized over time, after the initial repair, by the absence of a functioning pulmonary valve, which causes regurgitation, and by progressive enlargement of the right ventricle and pulmonary arteries. Due to this pathological anatomy, available transcatheter valves are usually too small to be deployed in the enlarged right ventricular outflow tracts (RVOT). To avoid surgical valve replacement, an alternative consists in implanting a reducer prior to or in combination with a transcatheter valve. We describe a computational model to study the effect of a stented valve RVOT reducer on the hemodynamics in enlarged ventricular outflow tracts. To this aim, blood flow in the right ventricular outflow tract is modeled via the incompressible NavierStokes equations coupled to a simplified valve model, numerically solved with a standard finite element method and with a reduced order model based on Proper Orthogonal Decomposition (POD). Numerical simulations are based on a patient geometry obtained from medical imaging and boundary conditions tuned according to measurements of inlet flow rates and pressures. Different geometrical models of the reducer are built, varying its length and/or diameter, and compared with the initial devicefree state. Simulations thus investigate multiple device configurations and describe the effect of geometry on hemodynamics. Forces exerted on the valve and on the reducer are monitored, varying with geometrical parameters. Results support the thesis that the reducer does not introduce significant pressure gradients, as was found in animal experiments. Finally, we demonstrate how computational complexity can be reduced with POD. 
A. Caiazzo, G. Montecinos, L.O. Müller, E.M. Haacke, E.F. Toro, Computational haemodynamics in stenotic internal jugular veins, Journal of Mathematical Biology, 70 (2015), pp. 745772.
Abstract
Stenosis in internal jugular veins (IJVs) are frequently associated to pathological venous circulation and insufficient cerebral blood drainage. In this work, we set up a computational framework to assess the relevance of IJV stenoses through numerical simulation, combining medical imaging, patientspecific data and a mathematical model for venous occlusions. Coupling a threedimensional (3D) description of blood flow in IJVs with a reduced onedimesional model (1D) for major intracranial veins, we are able to model different anatomical configurations, an aspect of importance to understand the impact of IJV stenosis in intracranial venous haemodynamics. We investigate several stenotic configurations in a physiologic patientspecific regime, quantifying the effect of the stenosis in terms of venous pressure increase and wall shear stress patterns. Simulation results are in qualitative agreement with reported pressure anomalies in pathological cases. Moreover, they demonstrate the potential of the proposed multiscale framework for individualbased studies and computeraided diagnosis. 
A. Caiazzo, I. RamisConde, Multiscale modeling of palisade formation in glioblastoma multiforme, Journal of Theoretical Biology, 383 (2015), pp. 145156.
Abstract
Palisades are characteristic tissue aberrations that arise in glioblastomas. Observation of palisades is considered as a clinical indicator of the transition from a noninvasive to an invasive tumour. In this article we propose a computational model to study the influence of genotypic and phenotypic heterogeneity in palisade formation. For this we produced three dimensional realistic simulations, based on a multiscale hybrid model, coupling the evolution of tumour cells and the oxygen diffusion in tissue, that depict the shape of palisades during its formation. Our results can be summarized as the following: (1) we show that cell heterogeneity is a crucial factor in palisade formation and tumour growth; (2) we present results that can explain the observed fact that recursive tumours are more malignant than primary tumours; and (3) the presented simulations can provide to clinicians and biologists for a better understanding of palisades 3D structure as well as glioblastomas growth dynamics 
C. Bertoglio, A. Caiazzo, A tangential regularization method for backflow stabilization in hemodynamics, Journal of Computational Physics, 261 (2014), pp. 162171.
Abstract
In computational simulations of fluid flows, instabilities at the Neumann boundaries may appear during backflow regime. It is widely accepted that this is due to the incoming energy at the boundary, coming from the convection term, which cannot be controlled when the velocity field is unknown. We propose a stabilized formulation based on a local regularization of the fluid velocity along the tangential directions on the Neumann boundaries. The stabilization term is proportional to the amount of backflow, and does not require any further assumption on the velocity profile. The perfomance of the method is assessed on a two and threedimensional Womersley flows, as well as considering a hemodynamic physiological regime in a patientspecific aortic geometry. 
A. Caiazzo, J. Mura, Multiscale modeling of weakly compressible elastic materials in harmonic regime and application to microscale structure estimation, Multiscale Modeling & Simulation. A SIAM Interdisciplinary Journal, 12 (2014), pp. 514537.
Abstract
This article is devoted to the modeling of elastic materials composed by an incompressible elastic matrix and small compressible gaseous inclusions, under a time harmonic excitation. In a biomedical context, this model describes the dynamics of a biological tissue (e.g. lung or liver) when wave analysis methods (such as Magnetic Resonance Elastography) are used to estimate tissue properties. Due to the multiscale nature of the problem, direct numerical simulations are prohibitive. We extend the homogenized model introduced in [Baffico, Grandmont, Maday, Osses, SIAM J. Mult. Mod. Sim., 7(1), 2008] to a time harmonic regime to describe the solidgas mixture from a macroscopic point of view in terms of an effective elasticity tensor. Furthermore, we derive and validate numerically analytical approximations for the effective elastic coefficients in terms of macroscopic parameters. This simplified description is used to to set up an efficient variational approach for the estimation of the tissue porosity, using the mechanical response to external harmonic excitations. 
TH.I. Seidman, O. Klein, Periodic solutions of isotone hybrid systems, Discrete and Continuous Dynamical Systems. Series B. A Journal Bridging Mathematics and Sciences, 18 (2013), pp. 483493.
Abstract
Suggested by conversations in 1991 (Mark Krasnosel'skiĭ and Aleksei Pokrovskiĭ with TIS), this paper generalizes earlier work (Krasnosel'skiĭPokrovskiĭ 1974) of theirs by defining a setting of hybrid systems with isotone switching rules for a partially ordered set of modes and then obtaining a periodicity result in that context. An application is given to a partial differential equation modeling calcium release and diffusion in cardiac cells. 
M. Grote, V. Palumberi, B. Wagner, A. Barbero, I. Martin, Dynamic formation of oriented patches in chondrocyte cell cultures, Journal of Mathematical Biology, 63 (2011), pp. 757777.
Abstract
Growth factors have a significant impact not only on the growth dynamics but also on the phenotype of chondrocytes (Barbero et al. , J. Cell. Phys. 204, pp. 830838, 2005). In particular, as chondrocyte populations approach confluence, the cells tend to align and form coherent patches. Starting from a mathematical model for fibroblast populations at equilibrium (Mogilner et al., Physica D 89, pp. 346367, 1996), a dynamic continuum model with logistic growth is developed. Both linear stability analysis and numerical solutions of the timedependent nonlinear integropartial differential equation are used to identify the key parameters that lead to pattern formation in the model. The numerical results are compared quantitatively to experimental data by extracting statistical information on orientation, density and patch size through Gabor filters. 
A. Barbero, V. Palumberi, B. Wagner, R. Sader, M. Grote, I. Martin, Experimental and mathematical study of the influence of growth factors and the kinetics of adult human articular chondrocytes, Journal of Cellular Physiology, 204 (2005), pp. 830838.
Preprints, Reports, Technical Reports

C. Bertoglio, A. Caiazzo, Y. Bazilevs, M. Braack, M. EsmailyMoghadam, V. Gravemeier, A.L. Marsden, O. Pironneau, I.E. VignonClementel, W.A. Wall, Benchmark problems for numerical treatment of backflow at open boundaries, Preprint no. 2372, WIAS, Berlin, 2017, DOI 10.20347/WIAS.PREPRINT.2372 .
Abstract, PDF (3076 kByte)
In computational fluid dynamics, incoming velocity at open boundaries, or backflow, often yields to unphysical instabilities already for moderate Reynolds numbers. Several treatments to overcome these backflow instabilities have been proposed in the literature. However, these approaches have not yet been compared in detail in terms of accuracy in different physiological regimes, in particular due to the difficulty to generate stable reference solutions apart from analytical forms. In this work, we present a set of benchmark problems in order to compare different methods in different backflow regimes (with a full reversal flow and with propagating vortices after a stenosis). The examples are implemented in FreeFem++ and the source code is openly available, making them a solid basis for future method developments.
Talks, Poster

K. Papafitsoros, Generating structure nonsmooth priors for image reconstruction, Young Researchers in Imaging Seminars, March 20  27, 2019, Henri Poincaré Institute, Paris, France, March 27, 2019.

J.A. Brüggemann, Pathfollowing methods for a class of elliptic obstacletype quasivariational problems with integral constraints, 23rd International Symposium on Mathematical Programming (ISMP2018), Session 370 ``Variational Analysis 4'', July 1  6, 2018, Bordeaux, France, July 2, 2018.

A. Caiazzo, Mathematical modeling and simulations of geothermal reservoirs, Virtual Physiological Human Conference (VPH2018), September 5  7, 2018, University of Zaragoza, Spain, September 6, 2018.

A. Caiazzo, Robust open boundary conditions and efficient data assimilation in multiscale hemodynamics, International Symposium ``Modeling, Simulation and Optimization of the Cardiovascular System'', October 22  24, 2018, Universität Magdeburg, October 22, 2018.

A. Caiazzo, Towards the personalization of (1D) bloodflow simulations, University of Amsterdam, Computational Science Lab, Netherlands, September 21, 2018.

M. Hintermüller, M. Holler, K. Papafitsoros, A function space framework for structural total variation regularization in inverse problems, MIA 2018  Mathematics and Image Analysis, HumboldtUniversität zu Berlin, January 15  17, 2018.

K. Papafitsoros, A function space framework for structural total variation regularization with applications in inverse problems, LAWOC 2018, VI Latin American Workshop on Optimization and Control, September 3  7, 2018, Quito, Ecuador, September 4, 2018.

A. Caiazzo, Estimation of cardiovascular system parameters from real data, 2nd Leibniz MMS Days 2017, February 22  23, 2017, Technische Informationsbibliothek, Hannover, February 22, 2017.

A. Caiazzo, Homogenization methods for weakly compressible elastic materials forward and inverse problem, Workshop on Numerical Inverse and Stochastic Homogenization, February 13  17, 2017, Universität Bonn, Hausdorff Research Institute for Mathematics, February 17, 2017.

A. Caiazzo, A comparative study of backflow stabilization methods, 7th European Congress of Mathematics (7ECM), July 18  22, 2016, Technische Universität Berlin, Berlin, July 19, 2016.

A. Caiazzo, Backflow stabilization methods for open boundaries, ChristianAlbrechtsUniversität zu Kiel, Angewandte Mathematik, Kiel, May 19, 2016.

A. Caiazzo, Multiscale modeling of weakly compressible elastic materials in harmonic regime, Rheinische FriedrichWilhelmsUniversität Bonn, Institut für Numerische Simulation, Bonn, May 21, 2015.
Mathematical Context
 Analysis of Partial Differential Equations and Evolutionary Equations
 Multiscale Modeling and Asymptotic Analysis
 Numerical Methods for PDEs with Stochastic Data
 Numerical methods for coupled systems in computational fluid dynamics
 Optimal control of partial differential equations and nonlinear optimization
 Systems of partial differential equations: modeling, numerical analysis and simulation
 Variational methods