Hello,
I had some questions regarding applying svFSI to create a moving, deformable model. I am working with time-series data, where each frame in the series captures an image of the aorta at that moment in time. If I want to generate a deformable, moving wall geometric model, how would I go about this? I'm thinking of creating meshes at different time points in the cardiac cycle. But then, how would I connect these meshes together to create one animated model that shows the walls of the aorta pulsating?
I saw the documentation/source code for this paper, https://asmedigitalcollection.asme.org/ ... mage-Based
but I'm a little confused. I see that they created a mesh at end diastole, and then used interpolation.py to create meshes at different points in the cardiac cycle. However, is this interpolation.py script only applicable to models of the Left ventricle? Additionally, how did they compose each generated mesh into a singular dynamic model?
Thank you for taking the time to read!
Generating a model with moving boundaries
-
Joanne Sarsam
- Posts: 34
- Joined: Tue Oct 25, 2022 11:44 am
-
Vijay Vedula
- Posts: 63
- Joined: Mon Feb 09, 2015 1:27 pm
Re: Generating a model with moving boundaries
Hi Joanne,
Usually, you need to perform registration, either a point-set registration or image-based registration, to extract nodal displacements and impose them to perform a prescribed-wall motion FSI simulation. Documentation on setting up such a simulation is provided here.
Reg. the reference you shared, I believe the scripts are applicable to the left ventricle primarily. But the registration methods employed are general and could be applied to modeling the aorta as well. You can find more details here. However, care should be taken to ensure smooth deformations around the branches of the aorta if you plan to include them.
-- Vijay
Usually, you need to perform registration, either a point-set registration or image-based registration, to extract nodal displacements and impose them to perform a prescribed-wall motion FSI simulation. Documentation on setting up such a simulation is provided here.
Reg. the reference you shared, I believe the scripts are applicable to the left ventricle primarily. But the registration methods employed are general and could be applied to modeling the aorta as well. You can find more details here. However, care should be taken to ensure smooth deformations around the branches of the aorta if you plan to include them.
-- Vijay
-
Joanne Sarsam
- Posts: 34
- Joined: Tue Oct 25, 2022 11:44 am
Re: Generating a model with moving boundaries
Hi Vijay,
Thank you for your response. If I create a mesh for, hypothetically, 4 different time frames over the course of a cardiac cycle, would it be possible to use the interpolation script linked here: https://github.com/SimVascular/SimVascu ... olation.py to interpolate the meshes at all points in the cardiac cycle (with appropriate time resolution), and then use all the interpolated meshes to define the motion .dat file needed for the prescribed motion simulation?
I guess what I'm getting at is, is it necessary to do the registration step if I have access to time series data? Or can I just conduct interpolation? If it is necessary, what is it's purpose?
Thank you,
Joanne
Thank you for your response. If I create a mesh for, hypothetically, 4 different time frames over the course of a cardiac cycle, would it be possible to use the interpolation script linked here: https://github.com/SimVascular/SimVascu ... olation.py to interpolate the meshes at all points in the cardiac cycle (with appropriate time resolution), and then use all the interpolated meshes to define the motion .dat file needed for the prescribed motion simulation?
I guess what I'm getting at is, is it necessary to do the registration step if I have access to time series data? Or can I just conduct interpolation? If it is necessary, what is it's purpose?
Thank you,
Joanne