OpenLB UseCase B
In UseCase A, we demonstrated how to create a simulation based on a geometry in a .stl file. This time, we will explain how to achieve the same result using measured data, including segmentation and extraction of the isosurface.
The starting point for this use case is measured data, such as intensity values from an MRI scan. Using this data, the Random Walker Segmentation technique can be employed to generate a segmented volume. Subsequently, the Isosurface Extraction processor extracts the geometry of the relevant data, which is then used for further analysis in the simulation.
Open the workspace for use-case B (segmentation).
Upon entering the application mode, you will see the following:
-
Input:
Load the dataset. -
Data Seetings:
Utilizes the VolumeListAggregate processor from Preprocessing Flow Data to aggregate different timesteps of the same measurement into 4D data. -
2D Rendering:
Configure the transfer function. -
Random Walker:
Initiate the random walker segmentation, as described in Random Walker Segmentation. -
Output:
Save the created segmentation in various file formats.
This process works as explained in Isosurface Extraction.
Open the workspace for use-case B (simulation).
Upon entering the application mode, you will see the following:
The processors used here are identical to those in UseCase A
After configuring the parameters, you can initiate and visualize the simulation. An example is shown here:
The possible interactions with the simulation results are the same as those in UseCase A
It is essential to note that the VolumeListAggregate processor assumes that the position of the examined volume remains relatively stable over time. For further explanation, please refer to Preprocessing Flow Data.