First time here? Checkout the FAQ!
x
0 votes
by (130 points)

I'm trying to use "tuneCV" in order to tune intra- and extracellular conductivities based on given CVs. However, I'm experiencing some issues.

First of all, I tried to run the simulation as in the tutorial "02_EP_tissue/03A_study_prep_tuneCV", but the results I obtain do not match those reported on the website. For example, when running:

./run.py --velocity 0.6 --converge 1

I obtain the following output:

Conduction velocity: 0.6168 m/s [gi=0.1740, ge=0.6250, gm=0.1361]
Conduction velocity: 0.5987 m/s [gi=0.1646, ge=0.5914, gm=0.1288]
Conduction velocity: 0.6001 m/s [gi=0.1653, ge=0.5939, gm=0.1293]

instead of:

Conduction velocity: 0.6398 m/s [gi=0.1740, ge=0.6250, gm=0.1361]
Conduction velocity: 0.6006 m/s [gi=0.1530, ge=0.5497, gm=0.1197]
Conduction velocity: 0.6001 m/s [gi=0.1527, ge=0.5486, gm=0.1195]. 

The difference is not massive, but I'm wondering whether to trust my results or not. Are these differences due to an update in the code?

I also noticed that the proximal tissue does not activate if the velocity is increased above a certain value (e.g., --velocity=1.1, if using the TT2 model). However, I'd need to compute the conductivities corresponding to the CV in a fast endocardial conduction layer (i.e., > 2 m/s). I tried to change both the spatial resolution and the integration time step, but so far with no success. Are there other parameters I should modify, such as the solver tolerance or the maximum number of iterations?

1 Answer

+1 vote
by (220 points)

There are many things that could produce such a small change, but those difference are not big enough to have any major relevance. 

For your fast conduction velocities problem, You could try to increase the strength of the stimuli, some higher conductivities required stronger stimuli to propagate. The parameter is called "--stimS" if you run it using the tutorial run.py file. you would have to add it between lines 350 and 385, (the default value is 250 uA/cm^2) 

by (130 points)
Thanks a lot! As suggested, I managed to compute the conductivities for larger values of CV by increasing the stimulus strength.
Welcome to openCARP Q&A. Ask questions and receive answers from other members of the community. For best support, please use appropriate TAGS!
architecture, carputils, documentation, experiments, installation-containers-packages, limpet, slimfem, website, governance
...