I have enabled reslice driver for tracking of a needle. The reslice rendering performance is quite good when I am navigating on only once slice at a time. However when I set the slab thickness to around 75 and have a spacing fraction of 5 I see a huge slowdown to about 1fps. Is this normal? I am seeing that my computer is being CPU limited. I have a i7 with 1.9ghz speed.
Yes, I would say that sounds normal. You might want to change to using GPU volume rendering instead (unfortunately it’s a very different pipeline so you’d probably need to change the way you do things).
If you want to limit the slab to a certain thickness then set the volume rendering clipping ROI’s Z size to the desired thickness and then apply the SlicerToRAS transform of the slice view to this ROI.
Thanks for the responses. I used the 3d view and followed the steps as Iassoan linked with some tweaks for the scan I have. It does seem to be much more responsive.
To set the view I set the camera focal point to the center of my volume and then I offset the camera position to a position above the focal point to get a Coronal image of the scan. This seems to work ok. Although the FOV isnt currently applicable to an volume since I set the distance manually.
Im a little confused about the last suggestion
I’m not using a slice view. Are you suggesting that I can place a volume render in a slice view window?
If the axes of the thick slab is aligned with anatomical axes then you don’t need a transform, you can just set the center and size of the ROI to define the slab.
If you want to show an oblique slab then you need to create a transform to position/orient your ROI node. If you use slice views, you can set this transform by copying the position/orientation of an oblique slice view.
I understand I need to get a transform if I want a arbitrary slice angle/oblique view. What I dont understand is how to get a volume render to display in a slice view. Is that what you are suggesting?
I’m not suggesting to display a volume rendered image in a slice view. If you need that then you need to implement it yourself (grab the image from the 3D view, put it into a single-slice volume, and show that volume in a slice view).