Calculating Short & Long Axis of Neurons from Z-stacked Images

[Kalen K]

I'm fairly new to using ImageJ, but familiar with using computers, software,
and other similar things.*

Background Info:*
Imagine you have a cell in 3D in the shape of a pear
(ellipsoidal<http://en.wikipedia.org/wiki/Ellipsoid>)
resting at a slight angle. There will be a short and long diameter also
called semi-minor <http://en.wikipedia.org/wiki/Semi-minor_axis> and
semi-major <http://en.wikipedia.org/wiki/Semi-major_axis> axis. If we assume
a cell is pear shaped the long axis will be angled longitudinal from the top
corner to the bottom, while the short axis will be transversely in the
middle.

*Task:*
To find the major and minor axis of the cell body of ~300 retinal ganglion
cells from z-stack confocal images.

*Some Ideas So Far:*
*1)* To measure the "transverse diameter," I would navigate to the z-slice
with the largest diameter and trace the cell body manually, and measure the
max & min feret diameter. To measure the "longitudinal diameter" I would
calculate the number of z-slices that the cell occupies and multiply by the
z-thickness to find the "height" of the cell. However, this assumes the
diameter is perpendicular and not slanted at some angle. Also, relying on
the z-thickness might be inaccurate, since fluorescent images tend to blur
along the z axis. Hence the use of the point spread function.

It would be nice if there was an easy to to select the cell across multiple
z-slices and find the max & min feret diameter across 3 dimensions...

*2)* Use the Threshold to define the cell and utilize Analyze Particles
along with the ROI Manager as described here:
http://n2.nabble.com/multiple-ROI-tp2371815p2371886.html

However, I am a little unsure how to put that all together and actual use it
towards my application.

*3)* To measure the "longitudinal diameter" someone suggested using the
Volume Viewer plugin to rotate and orient the cell longitudinally such that
the longest diameter is clearly shown. Then select "save image" and make my
measurement on the newly acquired image. As described here:
http://n2.nabble.com/vessel-diameter-tp633093p633094.html

However, I found the Volume Viewer to be very difficult to use in orienting
the cell into the correct position...


Any input or suggestions would be greatly appreciated.

Thank you,

Kalen

[Thomas Boudier]

Hi,

You can use 3D objects counter and the 3D axes plugin to do that, there
is a fitting by an 3D ellipsoid and you get the values you want.

http://imagejdocu.tudor.lu/doku.php?id=plugin:analysis:3d_object_counter:start&rev=1252653791

http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start

Best,

Thomas


Kalen K a écrit :

> I'm fairly new to using ImageJ, but familiar with using computers, software,
> and other similar things.*
>
> Background Info:*
> Imagine you have a cell in 3D in the shape of a pear
> (ellipsoidal<http://en.wikipedia.org/wiki/Ellipsoid>)
> resting at a slight angle. There will be a short and long diameter also
> called semi-minor <http://en.wikipedia.org/wiki/Semi-minor_axis> and
> semi-major <http://en.wikipedia.org/wiki/Semi-major_axis> axis. If we assume
> a cell is pear shaped the long axis will be angled longitudinal from the top
> corner to the bottom, while the short axis will be transversely in the
> middle.
>
> *Task:*
> To find the major and minor axis of the cell body of ~300 retinal ganglion
> cells from z-stack confocal images.
>
> *Some Ideas So Far:*
> *1)* To measure the "transverse diameter," I would navigate to the z-slice
> with the largest diameter and trace the cell body manually, and measure the
> max & min feret diameter. To measure the "longitudinal diameter" I would
> calculate the number of z-slices that the cell occupies and multiply by the
> z-thickness to find the "height" of the cell. However, this assumes the
> diameter is perpendicular and not slanted at some angle. Also, relying on
> the z-thickness might be inaccurate, since fluorescent images tend to blur
> along the z axis. Hence the use of the point spread function.
>
> It would be nice if there was an easy to to select the cell across multiple
> z-slices and find the max & min feret diameter across 3 dimensions...
>
> *2)* Use the Threshold to define the cell and utilize Analyze Particles
> along with the ROI Manager as described here:
> http://n2.nabble.com/multiple-ROI-tp2371815p2371886.html
>
> However, I am a little unsure how to put that all together and actual use it
> towards my application.
>
> *3)* To measure the "longitudinal diameter" someone suggested using the
> Volume Viewer plugin to rotate and orient the cell longitudinally such that
> the longest diameter is clearly shown. Then select "save image" and make my
> measurement on the newly acquired image. As described here:
> http://n2.nabble.com/vessel-diameter-tp633093p633094.html
>
> However, I found the Volume Viewer to be very difficult to use in orienting
> the cell into the correct position...
>
>
> Any input or suggestions would be greatly appreciated.
>
> Thank you,
>
> Kalen
>
>

--
   /**********************************************************/
      Thomas Boudier, MCU Université Pierre et Marie Curie,
      IFR 83. Bat B 7ème étage, porte 709, Jussieu.
      Tel : [phone]  Fax : [phone]
/*******************************************************/