Experimental mechanostimulation of biologic samples serves to understand fundamental processes in mechanobiology. Unfortunately, present techniques do not allow contact-free investigation of material characteristics. We developed a mechanostimulator that functions under the assumption that the counterforce generated by a biologic sample placed on a carrier membrane (test-on-carrier) is extractable from the total signal. As material characteristics of biologic samples are poorly defined, we substituted them by polyurethane and latex membranes. Based on the knowledge of the previously measured carrier membrane's compliance, we calculated the test membrane's compliance from the pressure-volume relationship inside the bioreactor. We hypothesized that knowing the carrier membranes' compliance allows calculation of the sample's compliance. Compliances of latex membranes were larger than those of polyurethane ones (p < 0.05). The average differences between calculated and directly determined compliances were 3.9% +/- 2.7% and 6.7% +/- 4.8% (mean +/- SD) under static and dynamic mechanostimulation, respectively (p > 0.05). The variability of compliance values was several fold larger (p < 0.01) for latex-on-polyurethane compared to polyurethane-on-latex combinations. There was high correlation between Young's modulus as determined from static compliance estimation and by uniaxial stretching in a tension-testing machine (r(2) = 0.987, p < 0.001). In conclusion, our mechanostimulator allowed reliable contact-free determination of material characteristics and may thus be suitable for contact-free assessment of biologic samples. The compliance of the carrier membrane should be at least two times larger than the compliance of the tested material.

No abstract

Using conventional methods, three-dimensional imaging of the lung is challenging because of the low contrast between air and tissue and the large differences in dimensions between various pulmonary structures. The small distal airway structures and the high air-to-tissue ratio of lung tissue require an imaging technique which reliably discriminates between air and water. The objective of this study was to assess whether neutron computed tomography would satisfy such a requirement. This method utilizes the unique characteristic of neutrons of directly interacting with the atomic nucleus rather than being scattered by the atomic shell. Neutron computed tomography was tested in rats and allowed differentiation of larger lung structures (e.g., lobes) and distal airways. Airways could be identified reliably down to the sixth bronchial generation, in some cases even down to the tenth generation. The lung could be stabilized for sufficiently long exposure times to achieve an image resolution of 50-60 µm, which is the current physical resolution limit of the neutron computed tomography facility. Neutron computed tomography allowed excellent lung imaging without the need for additional tissue preparation or contrast media. The enhanced structural resolution obtained by applying this new research technique may improve understanding of lung physiology and respiratory therapy.

No abstract

Abstract. This paper is concerned with coupled problems in the human respiratory system with emphasis on mechanical ventilation. We focus on the modeling aspects of pulmonary alveoli and the lower airways against the background of acute lung diseases. In this connection occurring stresses and strains are of substantial interest.For the first generations of the bronchial tree, a geometry based on human computer tomography scans obtained from in-vivo experiments is employed. The deformability of airway walls is taken into account to study airflow structures and airway wall stresses for a number of different scenarios. Therefore we carried out fluid-structure interaction (FSI) simulations under transient incompressible flow conditions. Both models for healthy and diseased lungs are studied under normal breathing as well as under mechanical ventilation.Our alveolar model is based on three-dimensional artificial random geometries generated with the help of a new labyrinthine algorithm ensuring preservation of overall minimal mean pathlength. A polyconvex hyperelastic material model incorporating general histologic information is employed to realistically describe alveolar parenchymal tissue properties. The influence of surface-active agents (the so-called surfactant) on the overall mechanical behavior of pulmonary alveoli is investigated. For this purpose an adsorption-limited model relating surface stresses to the interfacial concentration of surfactant is used.