Bugs in Orthogonal Views

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Bugs in Orthogonal Views

Stein Rørvik
I work with micro CT stacks where the start slice is the bottom of the object.
The convention in ImageJ is that the start slice is the top of the object.
I therefore select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
This works well, but I have encountered some bugs:

Try this:

1) select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
2) Open the T1 Head sample stack.
3) Select Image - Stacks - Orthogonal Views
4) Say we are interested in the right eye. Scroll to slice 43 and click at the eye.
   The ZY and XZ images are updated correctly.
5) Click the left eye in the XZ image. The stack and ZY images are updated correctly.
6) Click the left eye in the ZY image. The opposite Z position is selected instead.
7) Close Orthogonal Views (by selecting Orthogonal Views again, Ctrl-Shift-H).
8) Reopen Orthogonal Views. The XZ and ZY images are correct. The lines are correct.
   But the slice number of the XY stack is wrong.
   It is set to the middle slice (nSlices/2) so the slice does not match the orthoview.

Another issue is that when "Rotate YZ" and "Flip XZ" is active,
the Z coordinate logically increases from the bottom and up along the Y direction.
However, the Y coordinate in the images increase from top to bottom.
The InverseY flag should therefore have been set for the ZY and XZ images in this case.

That is, activating "Rotate YZ" should set the InverseY flag to true for the ZY image,
and activating "Flip XZ" should set InverseY flag to true for the XZ image.
Then all coordinates would correspond to the same number in all views.

So the bugs are as follows:
- When "Rotate YZ" is active, the selected Y position is inverted in the ZY image
- When Orthogonal Views is reopened, the slice number is wrongly set to nSlices/2
- The Y coordinate should have been inverted in the ZY image when "Rotate YZ" is active
- The Y coordinate should have been inverted in the XZ image when "Flip XZ" is active

I am using daily build ImageJ 1.52j with Java 1.6 on Windows 7/64-bit.


Stein

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Bugs in Orthogonal Views

Wayne Rasband-2
> On Dec 10, 2018, at 6:36 AM, Stein Rørvik <[hidden email]> wrote:
>
> I work with micro CT stacks where the start slice is the bottom of the object.
> The convention in ImageJ is that the start slice is the top of the object.
> I therefore select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
> This works well, but I have encountered some bugs:

"Orthogonal Views” should work better in the latest daily build (1.52j24) when "Rotate YZ" and "Flip XZ” are enabled. The daily build also fixes a bug that sometimes caused "Orthogonal Views” to freeze and it re-enables dragging in the XZ and YZ views.

-wayne


> Try this:
>
> 1) select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
> 2) Open the T1 Head sample stack.
> 3) Select Image - Stacks - Orthogonal Views
> 4) Say we are interested in the right eye. Scroll to slice 43 and click at the eye.
>   The ZY and XZ images are updated correctly.
> 5) Click the left eye in the XZ image. The stack and ZY images are updated correctly.
> 6) Click the left eye in the ZY image. The opposite Z position is selected instead.
> 7) Close Orthogonal Views (by selecting Orthogonal Views again, Ctrl-Shift-H).
> 8) Reopen Orthogonal Views. The XZ and ZY images are correct. The lines are correct.
>   But the slice number of the XY stack is wrong.
>   It is set to the middle slice (nSlices/2) so the slice does not match the orthoview.
>
> Another issue is that when "Rotate YZ" and "Flip XZ" is active,
> the Z coordinate logically increases from the bottom and up along the Y direction.
> However, the Y coordinate in the images increase from top to bottom.
> The InverseY flag should therefore have been set for the ZY and XZ images in this case.
>
> That is, activating "Rotate YZ" should set the InverseY flag to true for the ZY image,
> and activating "Flip XZ" should set InverseY flag to true for the XZ image.
> Then all coordinates would correspond to the same number in all views.
>
> So the bugs are as follows:
> - When "Rotate YZ" is active, the selected Y position is inverted in the ZY image
> - When Orthogonal Views is reopened, the slice number is wrongly set to nSlices/2
> - The Y coordinate should have been inverted in the ZY image when "Rotate YZ" is active
> - The Y coordinate should have been inverted in the XZ image when "Flip XZ" is active
>
> I am using daily build ImageJ 1.52j with Java 1.6 on Windows 7/64-bit.
>
>
> Stein

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Bugs in Orthogonal Views

Stein Rørvik
Thanks, it works fine now!

There is another minor issue I forgot to mention in my last mail:

It is often useful to duplicate the XZ and/or YZ images for documentation purposes.
The coordinate lines are then presented as an overlay, which is nice, as it is then removable.
But if the stack has a density calibration, the density calibration does not apply to the XZ and YZ images.
If this was fixed as well, it would be perfect!

Stein

-----Original Message-----
From: ImageJ Interest Group <[hidden email]> On Behalf Of Wayne Rasband
Sent: 12. desember 2018 06:47
To: [hidden email]
Subject: Re: Bugs in Orthogonal Views

> On Dec 10, 2018, at 6:36 AM, Stein Rørvik <[hidden email]> wrote:
>
> I work with micro CT stacks where the start slice is the bottom of the object.
> The convention in ImageJ is that the start slice is the top of the object.
> I therefore select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
> This works well, but I have encountered some bugs:

"Orthogonal Views” should work better in the latest daily build (1.52j24) when "Rotate YZ" and "Flip XZ” are enabled. The daily build also fixes a bug that sometimes caused "Orthogonal Views” to freeze and it re-enables dragging in the XZ and YZ views.

-wayne


> Try this:
>
> 1) select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
> 2) Open the T1 Head sample stack.
> 3) Select Image - Stacks - Orthogonal Views
> 4) Say we are interested in the right eye. Scroll to slice 43 and click at the eye.
>   The ZY and XZ images are updated correctly.
> 5) Click the left eye in the XZ image. The stack and ZY images are updated correctly.
> 6) Click the left eye in the ZY image. The opposite Z position is selected instead.
> 7) Close Orthogonal Views (by selecting Orthogonal Views again, Ctrl-Shift-H).
> 8) Reopen Orthogonal Views. The XZ and ZY images are correct. The lines are correct.
>   But the slice number of the XY stack is wrong.
>   It is set to the middle slice (nSlices/2) so the slice does not match the orthoview.
>
> Another issue is that when "Rotate YZ" and "Flip XZ" is active, the Z
> coordinate logically increases from the bottom and up along the Y direction.
> However, the Y coordinate in the images increase from top to bottom.
> The InverseY flag should therefore have been set for the ZY and XZ images in this case.
>
> That is, activating "Rotate YZ" should set the InverseY flag to true
> for the ZY image, and activating "Flip XZ" should set InverseY flag to true for the XZ image.
> Then all coordinates would correspond to the same number in all views.
>
> So the bugs are as follows:
> - When "Rotate YZ" is active, the selected Y position is inverted in
> the ZY image
> - When Orthogonal Views is reopened, the slice number is wrongly set
> to nSlices/2
> - The Y coordinate should have been inverted in the ZY image when
> "Rotate YZ" is active
> - The Y coordinate should have been inverted in the XZ image when
> "Flip XZ" is active
>
> I am using daily build ImageJ 1.52j with Java 1.6 on Windows 7/64-bit.
>
>
> Stein

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html
Reply | Threaded
Open this post in threaded view
|

Re: Bugs in Orthogonal Views

Wayne Rasband-2
> On Dec 12, 2018, at 4:38 AM, Stein Rørvik <[hidden email]> wrote:
>
> Thanks, it works fine now!
>
> There is another minor issue I forgot to mention in my last mail:
>
> It is often useful to duplicate the XZ and/or YZ images for documentation purposes.
> The coordinate lines are then presented as an overlay, which is nice, as it is then removable.
> But if the stack has a density calibration, the density calibration does not apply to the XZ and YZ images.
> If this was fixed as well, it would be perfect!

This bug is fixed in today’s daily build (1.52j26).

-wayne


> -----Original Message-----
> From: ImageJ Interest Group <[hidden email]> On Behalf Of Wayne Rasband
> Sent: 12. desember 2018 06:47
> To: [hidden email]
> Subject: Re: Bugs in Orthogonal Views
>
>> On Dec 10, 2018, at 6:36 AM, Stein Rørvik <[hidden email]> wrote:
>>
>> I work with micro CT stacks where the start slice is the bottom of the object.
>> The convention in ImageJ is that the start slice is the top of the object.
>> I therefore select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
>> This works well, but I have encountered some bugs:
>
> "Orthogonal Views” should work better in the latest daily build (1.52j24) when "Rotate YZ" and "Flip XZ” are enabled. The daily build also fixes a bug that sometimes caused "Orthogonal Views” to freeze and it re-enables dragging in the XZ and YZ views.
>
> -wayne
>
>
>> Try this:
>>
>> 1) select Options - Dicom Options and check both "Rotate YZ" and "Flip XZ".
>> 2) Open the T1 Head sample stack.
>> 3) Select Image - Stacks - Orthogonal Views
>> 4) Say we are interested in the right eye. Scroll to slice 43 and click at the eye.
>>  The ZY and XZ images are updated correctly.
>> 5) Click the left eye in the XZ image. The stack and ZY images are updated correctly.
>> 6) Click the left eye in the ZY image. The opposite Z position is selected instead.
>> 7) Close Orthogonal Views (by selecting Orthogonal Views again, Ctrl-Shift-H).
>> 8) Reopen Orthogonal Views. The XZ and ZY images are correct. The lines are correct.
>>  But the slice number of the XY stack is wrong.
>>  It is set to the middle slice (nSlices/2) so the slice does not match the orthoview.
>>
>> Another issue is that when "Rotate YZ" and "Flip XZ" is active, the Z
>> coordinate logically increases from the bottom and up along the Y direction.
>> However, the Y coordinate in the images increase from top to bottom.
>> The InverseY flag should therefore have been set for the ZY and XZ images in this case.
>>
>> That is, activating "Rotate YZ" should set the InverseY flag to true
>> for the ZY image, and activating "Flip XZ" should set InverseY flag to true for the XZ image.
>> Then all coordinates would correspond to the same number in all views.
>>
>> So the bugs are as follows:
>> - When "Rotate YZ" is active, the selected Y position is inverted in
>> the ZY image
>> - When Orthogonal Views is reopened, the slice number is wrongly set
>> to nSlices/2
>> - The Y coordinate should have been inverted in the ZY image when
>> "Rotate YZ" is active
>> - The Y coordinate should have been inverted in the XZ image when
>> "Flip XZ" is active
>>
>> I am using daily build ImageJ 1.52j with Java 1.6 on Windows 7/64-bit.
>>
>>
>> Stein

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html