How does SCT measure the convex area for its solidity calculation?

SCT
1

Hello ?

I have results of spinal cord metrics including solidity , AP , transverse diaemter etc of healthy controls. which is generated by SCT. I am gonna compare solidity of patients with certain pathology and these healthy controls. The problem is spinal cord MRI of patents includes 2D axial T2 weighted image. Thus I can not perform SCT tools for this 2D image. Instead I am gonna draw contour of axial spinal cord of patients. But I don’t know hot to measure convex area. In SCT process, definition of solidity is CSA/area convex. Can you share the algoristhm measuring “area convex” ?

Thank you in advance.

Sung Jun Ahn

2

Dear @ahn,

SCT’s code is public, so the solidity logic can be found here. In fact, SCT uses the solidity measure computed by the scikit-image.measure.regionprops utility.

In the scikit-image documentation, it says:

solidity
Ratio of pixels in the region to pixels of the convex hull image.

If I search the specific term “convex hull”, then I see the Convex Hull example page, and also the convex_hull_image function, which further links to the Binary image convex hull – algorithm notes post.

I hope this answers your question! :slight_smile:

Kind regards,
Joshua

3

Thank you Joshua.

It will be a great help !!!

Kind regards

Sung Jun

1 Like