Biomedical applications


Computational Biofluid Dynamics

Applications of large scale computations are developed to solve computationally demanding tasks in biofluid dynamics such as particle tracking in coupled fluid structure vascular domains or the hemodynamic assessment of numerous surgical options in vascular reconstructive surgery in a clinically useful time scale. High-order accurate c flow solutions in complex domains  allows the resolution of the full range of flow scales that develop for example in post stenotic jets that form in highly stenosed arterial conduits. In the field of magnetoheamodynamics with projects funded by the European Space Agency effects of strong static magnetic fields in the flow of blood in realistic arterial conduits are considered.

Quantitative Systems Pharmacology

Quantitative systems pharmacology (QSP) is a discipline that utilizes mathematical models and computing in order to describe biological systems, disease processes, and drug pharmacokinetics and pharmacodynamics. In case of pharmacokinetic and pharmacodynamic modeling, QSP can also allow the identification and quantification of several sources of variability (e.g., between- and within-subject, inter-occasion variability, etc.) in drug concentrations of the population under study. QSP modeling has numerous advantages such as incorporating unbalanced designs, modeling sparse data (e.g., only two or three samples per subject), and examining the role of patient-specific covariates (such as gender, age, body weight).

Modeling & Sumulation in drug development

Modeling and simulation (MS) approaches represent an integral part of drug development starting from in vitro testing up to all phases of the clinical procedure. The benefits of applying MS methods in drug development include the following: set the appropriate sampling scheme and estimate the necessary sample size, less human exposure in clinical trials, reduce the cost of drug development, optimize the clinical study design, allow for more rapid development procedures, early selection of the most appropriate chemical compounds and/or formulations, gain more information regarding the properties of the medicines, integrate in vitro and in vivo drug properties, individualize dose requirements. Today, the requirements of the European Medicines Agency (EMA) and the US-Food and Drug Administration (FDA) made the application of MS methods an essential step in drug development.

Modeling and Simulation of Mechanosensing and Cellular Locomotion

Cellular shape evolution of locomoting cells on deformable substrates is investigated by combining ideas of mechanosensing with advanced computational techniques for studying shape evolution,  using the level set method. We have developed new models for the evolution of locomoting fish epidermal keratocytes using a novel local active  mechanosensing hypothesis: cells contract their substrate and the lamellipodium evolves locally according to a local mechanosensing law, according to the local stress field caused by cellular contraction. The model proposed is tuned by iterative comaparison of simulation with observed behavior and captures a multitude of characteristic types of response. This activity sheds new light on the role of mechanosensing in cell migration and locomotion.

Neurophysiology of movement and attention

Spike-trains are recorded from single units of animals trained to execute movements with effectors such as the eye, the head or the arm and analyzed to gain insight into the movement variables coded by single neurons or populations of neurons. Particular emphasis is placed on: 1) the discharge pattern of neurons responsible for the control of the line of sight in the cat. 2) motor and visuomotor properties of the neurons in the ventral premotor area (area F5) of the macaque monkey. 3) functional organization and neuronal properties of the dorsal premotor cortex (area F2) of the macaque monkey. 4) characterization of the synaptic interactions between neurons with the help of electrophysiological and pharmacological methods.

Predictive Models in Biomedical Engineering and Epidemiology

Activities include a) predictive modeling for heart volumes which are used to perform virtual heart transplants, b) spatial statistical specifications for the growth of thrombus in abdominal aortic aneurysms, c) nonparametric functional classifiers for disease identification and d) spatio-temporal models for epidemics.