I have discovered that I am very lucky to work at the Institute of Biomedical Engineering at KIT. All my colleagues have been super friendly and helpful during my first weeks. They have even made some calls in German to help me fix my paperwork. Now almost all the bureaucracy is done, I am just waiting for my resident permit that is on its way.
This month, Carmen and Patricia, other ESRs from the consortium, started their work at KIT. That was amazing because I am not anymore the only new PhD at the department. They are both also Spanish speakers and we have become like a little family away from home.
The weather was colder this month but here you can see a picture of the beautiful colours that the sky has during fall. That is the view from my window at KIT
I did my first simulation in a very small piece of virtual tissue with two scars for 100 milliseconds. Here you can see a picture of my simulation. This piece of virtual heart is 10 mm x 5 mm x 0.5 mm. In red, you can see the tissue that is already activated. In blue, there is the tissue on the right side that has not been activated because the electrical wave has not propagated to that extreme of the tissue yet. The circular regions in blue represent scar tissue that is never activated by the electrical wave because it is not conductive
This video shows my other little success of the month: my first reentry is the same piece oh virtual heart. A reentry is an abnormal electrical wave that is perpetuated in time and is one of the main causes of atrial fibrillation (AF). This video captures only 3 seconds of this phenomenon but the computer needs almost 6 minutes to simulate this short time. Can you imagine how long would it take to make a simulation of the whole heart for a longer time??!!
I had to play with the conduction of the wave in different regions of the tissue. In most of the regions the wave is transmitted at a higher speed (i.e. normal tissue). In other regions the conduction is a bit slower (i.e sick tissue) and in the big region in the middle is close to zero because it represents scar tissue.
As I said before I was just playing around to understand the software and the computational model. For that reason, if you are an expert in these models you might notice that the velocity of propagation is a bit unrealistic. Nonetheless, I think It is a good start!
This felt like a nice achievement and it was very exciting to learn how to play with these 3D models. However, my objective is a lot more complicated than that. One of the most important goals of the Ph.D. at this early stage is to fully understand what the project is about. Very often, some of the ideas or preconceptions that you had about what you are supposed to do greatly change after reading some papers or having some conversations with the people involved in the field.
As I’ve already known, AF is a very common disease that does not immediately kill the patients like other heart diseases. However, it significantly increases the risk of mortality and decreases the quality of life in the long term. We know that AF is an irregular propagation of the electrical wave in the atria. But this irregularity is not fully understood by anybody, we only have a bunch of uncompleted theories about it. Hence, the available treatments for AF are still unsuccessful in many cases. Moreover, this irregular propagation has a high interpatient variability depending on the atria’s anatomy and other properties. Personalized computational models could be a valuable tool to assess the risk of developing AF in each patient.
A lot of mathematical models of electrical propagation in the heart have been proposed in the past. There are models at a cellular level, tissue level, or population level. They also replicate different areas of the heart such as ventricles or atria. In my case, I have to deal with models at the tissue level of the atria. The gold standard in this area is the biophysical models govern by the bidomain and monodomain equations. Using these models, a lot of phenomena in AF can be reproduced (i.e reentry). For example, the video that I showed you above was done using the monodomain equation. Unfortunately, they are computationally very expensive, sometimes requiring supercomputers to simulate a few seconds in the entire heart.
There is an alternative for the gold standard called the Eikonal models. They use the Eikonal equation, which is an approximate way of describing wave propagation. These models were originally used in optics but their applications nowadays include quantum mechanics, cardiac modeling, and many others. Eikonal models are a lot faster than the biophysical models but are incapable of simulating reentry, one of the most important phenomena in AF.
As you might guess, my task for the first year of the Ph.D. is to develop a mathematical model of the atria using the best aspects of each kind of model. It must have the speed of the Eikonal models to be suitable to be used at the hospital. At the same time, it should be able to reproduce reentry, as the biophysical models do. The temporary name of the model will be the “Reentrant Reaction Eikonal model” it might change in the future – I am open to suggestions. As you can imagine this is a very challenging goal but it will be very exciting and rewarding if my research produces good results. Let’s hope for the best…
Now I have finished my initial readings and know what I have to do. The next step is to define the mathematical framework of my model. If the equations that I will formulate make sense and are approved by my supervisors, I can go back to play with openCARP, this time to implement my own model. Moreover, I have been participating in several workshops organized for all the Ph.D. students of PersonalizeAF in the last couple of weeks. I will also prepare some tutorials for the master course Electromagnetics and Numerical Calculation of Fields at KIT. As you can see… November will be a very busy month for me. I will tell you how it goes in the next post!!
Until next time!
Cristian
Input your search keywords and press Enter.