IFAC PapersOnLine 52-26 (2019) 113–120 ScienceDirect ScienceDirect Available online at www.sciencedirect.com 2405-8963 © 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved. Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2019.12.245 1. INTRODUCTION Tissue engineered implants are a valuable contribution in the replacement of non-functional tissues and organs. So-called biohybrid implants consist of a textile scaffold, which is covered with in vitro grown connective tissue. Due to their components, biohybrid implants combine mechan- ical stability and biological compatibility (Sodhani et al., 2018). Among other applications they offer great potential in the area of artificial heart valves, because biohybrid heart valves are able to withstand mechanical stress while being highly biologically compatible. Furthermore, the liv- ing tissue which covers the scaffold is developed from the patient’s own cells (Flanagan et al., 2007). For this reason, biohybrid heart valves are able to grow and regenerate and therefore adapt to their surrounding. Correspondingly, a biohybrid heart valve is a complex system, which reacts to and interacts with its environment. Though there exist other approaches such as in situ tissue engineering (e.g. Bouten et al., 2018), the maturation process of a bio- hybrid heart valve is usually executed in a bioreactor. One potential bioreactor is described by Flanagan et al. (2007): While the heart valve is perfused by a pulsating flow of nutrition solution to imitate the cardiac cycle, the process can be manipulated by multiple actuating variables. Mechanical stimuli such as hydrostatic pressure and flow rate as well as biological stimuli through the composition of the nutrient solution can be applied. As an additional complexity, the maturation process is expected ⋆ This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 403043858. to be spatially distributed, meaning that the tissue growth properties vary across the heart valve. For a robust production of implantable biohybrid heart valves it is necessary to standardize the complex matu- ration process while considering the uniqueness of every implant. Therefore, the automation of the maturation process of biohybrid heart valves is required. The main challenge of the automated production of these artificial heart valves is the high degree of individuality of each liv- ing tissue and therefore each growing heart valve through e.g. individual speeds of growth. A model-based control scheme is a possible solution to this problem: Depending on whether the parameters vary during the maturation process, the patient specific model parameters can be identified either offline ahead or online during the process. In doing so, the model-based control is able to adapt to the individual heart valve. Thus, mathematical models are a great contribution to the understanding of this matura- tion process and the development of corresponding robust control algorithms. For the validation of such a model, experimental data are required, but experiments are time consuming and expensive: The maturation process of a biohybrid heart valve has a duration of three weeks (Mor- eira et al., 2016). Apart from this, due to the vulnerability of the growing tissue to contamination, the maturation process frequently fails to finish successfully. Consequently, only a low number of accomplished experiments can be performed. In addition, the only present possibility of a state analysis is a histological analysis of the heart valve, which destroys it. Therefore, such a snap-shot of the matu- rating heart valve results in a low amount of experimental Keywords: Distributed parameter systems, Tissue engineering, Numerical analysis, Mathematical models, Model-based experimental design, Sensitivity analysis Abstract: Model-based experimental designs minimize the experimental effort while maximiz- ing the amount of analyzable experimental data. In this paper, an initial mathematical model of the spatially distributed maturation process of tissue engineered heart valves is developed for a model-based experimental design. The model contains the state variables cell amount, collagen concentration and elastin concentration. Based on the developed model, a variance- based sensitivity analysis using Sobol’s method is performed. The results indicate that all model parameters influence the model solution, while the variance of the parameters related to cell diffusion comprises an exceeding influence. Consequently, corresponding experiments should especially focus on those parameters. * Institute of Automatic Control, RWTH Aachen University, Germany (e-mail: k.voss@irt.rwth-aachen.de) ** Institute of Applied Medical Engineering, RWTH Aachen University, Germany Kirsten Voß * Lorenz Pyta * Jonas Gesenhues * Petra Mela ** Thomas Schmitz-Rode ** Dirk Abel * Towards a Model-Based Experimental Design of the Maturation Process of Biohybrid Heart Valves ⋆ © 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.