The problem of excitation conduction from Line to cuboid

Question

Hello，

I am using OpenCARP for simulations and have encountered some issues. May you please provide some solutions?

The model I am using consists of two types of elements: Line (Ln) and tetrahedra (Tt). The model is as shown in the image, with the endpoint of the Line on the right connected to a vertex on the left of the cuboid. My goal is to stimulate the Line on the far left and have the excitation propagate to the cuboid on the right.

However, when I try to stimulate the Line by transmembrane current, I found that the excitation of Line on the left does not propagate to the right tetrahedral elements. I used "Time Series" in Meshalyzer to observe changes in transmembrane potential at the intersection point and found that the maximum value at the intersection point is -66. Therefore, I believe that there is no action potential generated at this location.

However, when I disconnect the Line from the cuboid and check the action potential at the original intersection point, as shown in the image, action potentials are generated after the disconnection.

Referring to previous Q&A from other users, I tried to connect the endpoint of the Line to about 10 vertices of cuboid, but the problem remains unsolved.

Can you please suggest what modifications I need to make to enable the excitation to propagate normally to the cuboid on the right?

Thank you in advance for your assistance!

Answer 1

Thank you for the clear description of your issue. To make it even easier to reproduce, adding a minimal working example can help for future issues.

What you describe is a classic situation of source sink mismatch. My guess would be that at the center of your line, you'll see action potentials in both cases (connected and disconnected) whereas at the end, the large sink formed by the cuboid leads to propagation failure at the end of the line.

This issue is very similar to the exit pathways in the sinus node driving the surrounding atrial tissue. The biological phenomenon and potentials ways to overcome it in modeling & simulation studies are for example described in this paper:  https://doi.org/10.1016/j.bpj.2022.10.020

Comments

Thank you very much for your response. The issue you pointed out regarding the source-sink mismatch is crucial.

Over the past period, I followed your advice and conducted experiments on small-sized cuboids (Tt) of various dimensions connected to large-sized cuboids (Tt). I validated the existence of the source-sink mismatch, and this mismatch issue can be overcomed through the design like referenced paper.

In my problem, through design a simplified model, I also found that by reducing the dimensions of the cuboids connected with Line (to overcome the source-sink mismatch issue), excitation can be transmitted from Line to the cuboid.

However, at the same time, I observed some changes in the action potential at the crosspoints (point 8, 9 in this fig(https://drive.google.com/file/d/1mgHN0FZQfBc0-cPDIv8oYUZVcyMzgM6d/view?usp=sharing)) of the Line to the cuboid. I speculate that the excitation on the cuboid is being transmitted back to Line. Is this a normal change?

Following your advice, I have packaged the code and mesh as example. Please download them from this link if needed.(https://drive.google.com/drive/folders/1nhf8Gi8G12VoMgfqOUSq9EW7W2-5lM0R?usp=drive_link)

Once again, thank you for your guidance!

---

Are you referring to the repolarizing dip right after the peak of the APs at points 8 and 9 that are present in the upper but not in the lower panel?
This should be caused by the currents that's drained from these elements to the cuboid to trigger cause depolarization and trigger an AP there. As soon as all neighboring elements are at the same potential the transmembrane voltage rises again in your line elements in line with the AP in the cuboid element.

---

Hi Hairui!

Thank you so much for making this question! I am dealing with the same problem as you. I was able to depolarize the cuboid from the line as you did but only in the situation the cuboid was extremely small (5um side and 5um resolution). For example, depolarization halted again when a small cuboid of 50 um size and 5 um resolution was used. Were you able to depolarize bigger cuboids from a single line?  

You can see my question here https://opencarp.org/q2a/1112/purkinje-on-opencarp if you are interested!

Maxi