Extra Line of Intersections in FIJI Sholl Analysis

imagej

#1

Hello,
I am a new user of FIJI, and I am trying to run a Sholl analysis of interneurons in the spinal cord dorsal horn of mice. I have a simple black and white image of the neuron, with no noise in the background, yet whenever I try to run Sholl the program counts an extra line of intersections extending down from the soma.
I’ve tried this with a number of similar images, and on a number of versions of FIJI.
I should also note that I’m using the Sholl Analysis feature built into FIJI (the one listed under “Analysis”- not the separate plugin offered by the Ghosh lab).
Is there any way to get rid of this line so these intersections are not counted in the Sholl Results table?


#2

Also having this issue


#3

I actually figured out how to get rid of it later- At least in the case of my images, I had thresholded them before putting them into ImageJ. I figured this was fine because the program ran Sholl without protest (usually if the image is not thresholded, if you try to click Sholl Analysis it will tell you to threshold first).
However, I actually needed to re-threshold it in ImageJ for the line to go away.

In certain cases I also ran into a related problem, which I am not sure if you are experiencing: In neurons that had a branch extending directly downward from the soma, there was a similar line of intersections within the black space of the neuron. To get rid of this, I wound up skeletonizing the image (Process>> Binary>> Skeletonize). This way, the neuron is reduced to a line drawing and there is no black space for extra intersections to go.
**Another note: I found that skeletonizing the image often caused the Sholl analysis to register many of the dendritic spines as branch intersections- to fix this I actually erased the spines from the image for the purposes of running Sholl.