open-discussion > Generating image for publication [Bioimage suite related]
Showing 1-4 of 4 posts
Display:
Results per page:
May 30, 2018  05:05 AM | Luis Araujo
Generating image for publication [Bioimage suite related]

Dear experts,

I have been working on an attempt to analyse SPECT data for the first time (previous questions at nitrc forum: 1, 2, 3 ). I used ISAS bioimage suite to calculate subtractions of inter and ictal SPECTs. Finally, imported them to SPM for the second level analysis (one sample T-test, cluster wise, no extended threshold).

Now I want to generate an image for my publication, using MRIcron is very hard with this type of data, as it has a lot of noise, causing it to be unusable. Which approach would you suggest for this case?

Thank you,
luis

‚Äč
May 30, 2018  05:05 AM | Larry Olson - Emory University
RE: Generating image for publication [Bioimage suite related]
It depends on the type of image you wish to show:
-MRIcron is actually excellent for extracting the VOI and displaying 2D images in a single plane or as orthogonal triplane images. I have found these steps helpful:
  • First, create an intensity filter of the perfusion image from the Draw menu. This will allow you to "extract" the volume of the hyper (or hypo) perfusion clusters.
  • Then mask the clusters in the perfusion image using the Draw menu. Open the image of the clusters, then open the intensity filter voi as a VOI in the Draw menu, then apply the mask and save the result.
  • Finally open the masked perfusion cluster as an overlay on your T1. I favor the actc LUT/color map to show these results
BioimageSuite's ISAS overlays are also really nice 2D representations but I do not find their 3D view quite as nice as:

-The "nicest" 3D representation can be done with MeVisLab's View3D.
  • Put the stripped T1 in as the main image,
  • then add the perfusion cluster you created with the MRIcron method as an isocontour overlay. I attached a typical MeVisLab network to show the setup.
  • You can clip the T1 in any plane to expose the deeper portions of the cluster or modify the transparency to do that.
Attachment: Capture.JPG
Jul 11, 2018  06:07 AM | Luis Araujo
RE: Generating image for publication [Bioimage suite related]
Thank you very much for the reply, it helped generate much better images.

I plan to use mricron for 2D images, how should I go about only filtering the significant hyperperfusion areas from my SPM second-level analysis?
Jul 11, 2018  07:07 AM | Larry Olson - Emory University
RE: Generating image for publication [Bioimage suite related]
-MRIcron is actually excellent for extracting the VOI and displaying 2D images in a single plane or as orthogonal triplane images. I have found these steps helpful if you choose MRIcron:
  • First, create an intensity filter of the perfusion image from the Draw menu. This will allow you to "extract" the volume of the hyper (or hypo) perfusion clusters.
  • Then mask the clusters in the perfusion image using the Draw menu. Open the image of the clusters, then open the intensity filter voi as a VOI in the Draw menu, then apply the mask and save the result.
  • Finally open the masked perfusion cluster as an overlay on your T1. I favor the actc LUT/color map to show these results
-BIS representations in the ISAS analysis are really nice too