Bone Volume Fraction Calculation

[Charison]

Hi All,

I am currently using the bone volume fraction option to calculate BV/TV on a sphere created by BoneJ. The measurements look fine to me. However when I calculate BV/TV on the inner cube I get a very different reading (higher) than what I get from the sphere. 

As such I was wondering if the bone volume fraction measurement from a sphere also measures the empty black spaces surrounding the sphere as part of the total volume?

Kind Regards

Charison

[Michael Doube]

Hi Charlson,

Did you draw an ROI on your sphere, or add the ROIs to the ROI Manager? If not, then the black pixels outside the sphere pixels will get counted as TV.

To restrict BV/TV measurement to a volume that is not the whole volume of the stack, you have to have ROIs in the ROI Manager and tell the plugin to use them (there's an option to tick).

Michael

Charison

--
You received this message because you are subscribed to the Google Groups "BoneJ Users and Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bonej-users-and-dev...@googlegroups.com.
To post to this group, send email to bonej-users-a...@googlegroups.com.
Visit this group at http://groups.google.com/group/bonej-users-and-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Charison]

Hi Michael, I did not. I think that is why most of us in the lab are getting lower values than what we expect. Some of us are now using the inner cubes to calculate our BV/TV and Tb.Sp.

On Monday, April 29, 2013 12:06:50 PM UTC+1, Michael Doube wrote:

Hi Charlson,

Did you draw an ROI on your sphere, or add the ROIs to the ROI Manager? If not, then the black pixels outside the sphere pixels will get counted as TV.

To restrict BV/TV measurement to a volume that is not the whole volume of the stack, you have to have ROIs in the ROI Manager and tell the plugin to use them (there's an option to tick).

Michael

On 29 April 2013 11:58, Charison <taych...@gmail.com> wrote:

Hi All,

I am currently using the bone volume fraction option to calculate BV/TV on a sphere created by BoneJ. The measurements look fine to me. However when I calculate BV/TV on the inner cube I get a very different reading (higher) than what I get from the sphere. 

As such I was wondering if the bone volume fraction measurement from a sphere also measures the empty black spaces surrounding the sphere as part of the total volume?

Kind Regards

Charison

--
You received this message because you are subscribed to the Google Groups "BoneJ Users and Developers" group.

To unsubscribe from this group and stop receiving emails from it, send an email to bonej-users-and-developers+unsub...@googlegroups.com.

[Michael Doube]

Hi Carison,

One analysis sequence might go like this (at least I think this is what I had in mind when putting these things together - it was a couple of years ago at least)

1) Open your image of e.g. a femoral head

2) place point ROIs on the surface of the head, each time hitting  [t] to add it to the ROI Manager

3) Run Plugins > BoneJ > Fit Sphere, clearing the ROI manager of the point ROIs and adding the sphere ROIs (set of circles) to the ROI Manager

4) Adjust the threshold (but don't apply it, it's sufficient to have the pixels you want analysed temporarily masked in 'red')

5) Run Plugins > BoneJ > Volume Fraction, Thickness, etc. selecting to restrict to ROI.

Note that you don't have to make an inner cube or sphere copy of the pixel data in this process, you simply restrict the analysis to the spherical ROI which is applied to the original pixel data.

Michael

To unsubscribe from this group and stop receiving emails from it, send an email to bonej-users-and-dev...@googlegroups.com.

[Richard Holder]

Hi all,

I've used BoneJ in the same way as my colleague to calculate BV/TV from a sphere I made in BoneJ. Unfortunately I did not add the sphere ROIs to the ROI Manager before doing this. Furthermore I have a dataset and a set of spheres from previous work, the creator of which did not use or save the sphere ROIs either.

Is it possible to get the correct numbers (calculated in just the spherical ROI) from the numbers calculated with the sphere and the 'black' space around it? I thought that, since the ratio of the volume of a cube to the volume of the largest sphere that fits within it would always be the same, I would be able to add this to a formula and use it to calculate the true BV/TV. Is that right? And would I be able to do something similar to other parameters which may be affected by overestimation of the total volume?

Failing that, is there anyway to obtain the sphere ROIs if you just have the sphere?

Best Wishes,

Richard

--

Richard L Holder

Fourth Year Medical Student

Imperial College School of Medicine

[Michael Doube]

Hi Richard,

Good thinking to subtract the difference between the sphere and the space around it from your TV then recalculate BV/TV with the same BV. It should work OK, provided that you know the padding (number of pixels between the sphere and side of the stack). That's probably the quickest way to do it because you could just run it as an equation in a spreadsheet in the cell beside BV and TV. Sphere TV would be something like:

stack edge length / 2 - padding = radius

then TV of sphere = 4/3 * PI * radius³

Make sure you correct your calculation for pixel spacing!

The end result should be a near constant ratio of cube volume to sphere volume with a slight adjustment for padding i.e. about (2r)³ : 4/3 * PI * r³ = 8r³ : 4PI/3 r³ = 2 : PI/3 = 1 : PI/6 (Please check that before using it!)

To reproduce the sphere ROIs you would have to run Fit Sphere again using point ROIs placed on your sphere image.

Michael

[Richard Holder]

Hi Michael,

I got a ratio of 1:Pi/6 too,; glad to know I got it right!

Would I be right in thinking that Connectivity Density would be affected too? If so, would I be be able to run Connectivity through a similar equation using the same ratio?

[Michael Doube]

Would I be right in thinking that Connectivity Density would be affected too? If so, would I be be able to run Connectivity through a similar equation using the same ratio?

Yes, Conn.D will be affected because the connectivity value is divided by a volume to get connections per unit volume. So if you used the whole stack you will be out by the same amount. Just use the TV you already calculated.

An additional point here though is that Conn.D contains an adjustment for the way the cut members on the edges of your data are not really cut in reality: in the specimen they joined on to something. BoneJ's implementation makes this adjustment by checking the faces of the stack, but it can't tell that your specimen is a sphere with lots of cut surfaces on it. I'm not sure how this will influence your results, but it's worth checking out to see if an inner cube gives you a similar value for Conn.D.