Interactive, Visual Simulation of a Spatio-Temporal Model of Gas Exchange in the Human Alveolus

In interdisciplinary fields such as systems biology, good communication between experimentalists and theorists is crucial for the success of a project. Theoretical modeling in physiology usually describes complex systems with many interdependencies. On one hand, these models have to be grounded on experimental data. On the other hand, experimenters must be able to understand the interdependent complexities of the theoretical model in order to interpret the model’s results in the physiological context. We promote interactive, visual simulations as an engaging way to present theoretical models in physiology and to make complex processes tangible. Based on a requirements analysis, we developed a new model for gas exchange in the human alveolus in combination with an interactive simulation software named Alvin. Alvin exceeds the current standard with its spatio-temporal resolution and a combination of visual and quantitative feedback. In Alvin, the course of the simulation can be traced in a three-dimensional rendering of an alveolus and dynamic plots. The user can interact by configuring essential model parameters. Alvin allows to run and compare multiple simulation instances simultaneously. We exemplified the use of Alvin for research by identifying unknown dependencies in published experimental data. Employing a detailed questionnaire, we showed the benefits of Alvin for education. We postulate that interactive, visual simulation of theoretical models, as we have implemented with Alvin on respiratory processes in the alveolus, can be of great help for communication between specialists and thereby advancing research.


S1.2 Implementation of Alvin
The application Alvin was built in Unity version 2020.1.16f1 (https://unity.com/). Unity is a platform for creating interactive real-time content. The mathematical model (see Section 3.1.1) provides the basis for the simulations. Parameter value changes in Alvin result in instantaneous updates of the visual output. All calculations are performed in the pressure unit mmHg. For the purpose of visualization, simulation time is slowed down by a factor of 40 compared to the gas exchange process in vivo.

S1.3 Three-dimensional visual model of an alveolus
The three-dimensional, mesh-based geometric model of an alveolus was created in Blender ® version 2.82 (https://www.blender.org/). The model resembles a real alveolus not only in its general appearance, but also in proportions. The tissue barrier, consisting of alveolar lining fluid, epithelial cells and connective tissue fibers are represented by a single layer in this model. This layer forms a truncated sphere with diameter, volume and surface area comparable to the corresponding values of an alveolus reported in the literature (Table S1). A network of hollow channels was modeled around this tissue layer, representing the capillary network enveloping the alveolus. The properties of the model network are chosen such that the relative magnitude of the volume and the surface area, as well as the relative radius and the length of the individual segments agree with the respective measured values from the literature (Table S1). From these morphometric features, a mean number of 52 capillaries surrounding an alveolus was derived ( Table 1). As the alveolar capillary network has no distinct capillaries, this value serves as a rough reference value. To determine the number of capillaries in our model, the network could be interpreted as parallel capillaries with cross-links. Our model contains 41 'parallel capillaries'. The cross-links between them contribute to gas exchange as well. To facilitate visualisation of the blood flow, a capillary was cut open longitudinally, thereby exposing the inside of the channel. In addition, the inflow and outflow of blood is indicated with the help of additional capillaries that connect to the capillary net.

S1.4 Simulation Output
In the graph panel, dynamic plots record the course of the simulation quantitatively. The plot "oxygen saturation along capillary" presents results obtained directly from the simulation calculations. The oxygen uptake, presented in another graph, is calculated assuming a standard amount of 270 · 10 6 hemoglobin molecules per erythrocyte (Pierigè et al., 2008). Considering the parameter values from the configuration menu, but independent of the rest of the simulation, the "oxygen dissociation curve" is calculated for a range of partial pressure values of oxygen.
If several simulation instances are active at the same time, the respective results are displayed together in the graphs, while only information about the selected instance is considered in the 3D visualisation in the center.
Two different prototypes of Alvin were implemented for the different use cases described in this work. One prototype was adapted for educational use (Section 3.3.2) such that the application has two levels of different complexity. In the first level, only one simulation instance can run at a time. The second level with full complexity is unlocked by an access code. The other prototype features additional readouts. "Membrane" diffusing capacity for oxygen DMO 2 and reaction half-time were required for model validation (Section 3.1.2). For the application example in research (Section 3.3.1), the output for diffusion capacity of the lung for oxygen was added. This second prototype is depicted in Figure 4 and available for download at https://go.uniwue.de/alvin.

S2.1 Questionnaire
The questionnaire was translated from the German original.

Demographics
In this section, we ask you to answer questions for general demographic information. These are relevant for a correct interpretation of your further answers.

Subject-specific exercises
In this group of questions, you will be given tasks that you can answer using the system. We ask you to discuss comments on the use of the app only in a joint round at the end of the event in order to minimize influencing the other participants. Now, familiarize yourself with the application. Look at how the graphs change in response to the controllers. Also observe how different disease patterns affect the values.
1. Which correlations between the course of the oxygen saturation curve ("Oxygen saturation along capillary") and the visualized simulation can you identify?
In this and the following blocks of questions, you will be given tasks to answer using the system. After each task (there are 3 tasks in total, each with subtasks), the answers will be discussed in plenary. We ask you to discuss comments on the use of the application only in a joint round at the end of the event in order to minimize influencing the other participants.
2. How does the oxygen dissociation curve change, when the body temperature rises to 40°C (fever)? 3. How does this affect the ability of hemoglobin to bind oxygen in the lungs? 4. How does it affect the ability of hemoglobin to deliver oxygen to tissues? Fever is normally accompanied by an increase in respiratory rate. By increasing the respiratory rate, the increased CO 2 produced by the increased metabolism during fever can be better exhaled. The partial pressure of CO 2 in the blood affects the ability of hemoglobin to bind O 2 .
5. By how many mmHg must the venous CO 2 partial pressure be lowered to achieve the same oxygen saturation at 40°C as at 37°?
The cruising altitude of passenger aircrafts is around 10 to 13 km. At this altitude, the partial pressure of oxygen is only between 30 and 44 mmHg. Therefore, the air pressure in the cabins of passenger aircrafts is artificially increased, but only to a level corresponding to the air pressure at about 2000-2500 m above sea level. Thus, an oxygen partial pressure of approx. 60 mmHg is achieved in venous blood. 6. What oxygen saturation does this correspond to? 7. At what alveolar pO 2 can a healthy person achieve this? 8. What oxygen saturation does the blood of a patient suffering from COPD reach at the same atmospheric pressure? 9. What happens to the oxygen saturation of a person suffering from COPD if he or she develops a fever during a flight?
In this block of questions, you will be given tasks to answer using the system. Please configure the application using the activation code provided in the lecture. We ask that you do not discuss comments on the use of the application until a joint round at the end of the course to minimize influencing the other participants.
10. Sometimes a lung has to be surgically removed due to a disease. What effects does this have on the oxygen saturation of the blood?
In tissues with very high metabolic rates, for example heavily used muscles, the CO 2 concentration can increase.
11. How does this affect the oxygen dissociation curve? 12. How does it affect oxygen uptake in the lungs and oxygen delivery in the tissues?
Start two simulation instances with the parameters for a healthy person. Increase the partial pressure of CO 2 in the arterial blood of one instance to 75 mmHg. Using the oxygen dissociation curves, measure the absorbed oxygen in the lungs and the delivered oxygen in the tissues.
13. At which blood pCO 2 is more oxygen available to the tissue? This phenomenon is called the Bohr effect.
Note: Exercises 14 to 17 were not covered in the educational case study (Section 3.3.2) due to time constraints.
Athletes, especially high-altitude mountaineers, can adapt to conditions at high altitude by training for longer periods at low oxygen partial pressure. This increases the diphosphoglycerate (DPG) concentration in the erythrocytes. Now, we would like to understand why this is beneficial. Start a simulation instance with the parameters for a healthy subject. First, reconstruct the conditions that exist when climbing at high altitudes: Decrease atmospheric pressure until alveolar pO 2 drops to a low value such as 40 mmHg. Arterial pO 2 is also reduced in these conditions. Set this to 30 mmHg.
14. What is the oxygen saturation of the blood? Duplicate the instance. Now, set the DPG concentration in one of the two instances to the maximum value (adjustment to high altitude).
15. What happens to the oxygen dissociation curve? 16. What happens to the oxygen saturation of the blood? 17. The effect you observed initially appears to be rather disadvantageous. Now, measure the oxygen saturation in the lungs and in the tissue in the respective graphs and determine the difference between these values. 121000 (Mercer et al., 1994) 150000 Capillary volume [µm 3 ] 808000 abstracted from (Gehr et al., 1978;Ochs et al., 2004) 787000 Capillary surface area [µm 2 ] 479000 abstracted from (Gehr et al., 1978;Ochs et al., 2004) 335000

TABLES AND FIGURES
Capillary radius [µm] 3.15 (Mühlfeld et al., 2010) 4.28 Capillary segment length [µm] 5.92 (Mühlfeld et al., 2010) 8.62 Table S1. The visual three-dimensional model of an alveolus was created in Blender ® . The size ratios were based on morphometric values from the literature. Figure S1. Participants were asked to solve 13 subject-specific exercises with the help of Alvin. Responses were scored 1 -correct, 2 -partially correct (e.g. subsequent faults), 3 -unclear to 4 -incorrect. Of the N = 73 participants, N = 31 were assigned to the group with previous knowledge level 1 (attendance of physiology lecture and / or knowledge from school or training). N = 34 participants were assigned to the group with previous knowledge level 2 (attendance of physiology lecture and additional literature).