ImageJ

Problem with "Orthogonal Views"

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

4 messages Options

Douglas Benn-2

Problem with "Orthogonal Views"

I wonder if anybody has the same problem with "Orthogonal Views"? When I do the following

run("Image Sequence...", "open=[C:\\CBCT research\\04\\ID001122012.dcm] number=525 starting=1 increment=1 scale=100 file=[] or=[] sort");
selectWindow("04");
run("Orthogonal Views");

It displays 3 windows - top left is axial slices, top right is saggital and bottom left coronal. However, the saggital images are flipped 90 degrees clockwise and the coronal 180 degrees.

I can correct the orientation by using

selectWindow("YZ 256");
run("Rotate 90 Degrees Left");
selectWindow("XZ 256");
run("Rotate... ", "angle=180 grid=1 interpolation=Bilinear");

However, as soon as I select the axial window to drag the planes cursor, the other windows flip again.

Does anybody know how to correct this?

Thank you.

Douglas

Dr Douglas K Benn, BDS,DDS, M.Phil., Ph.D., Dipl. Dental Radiology (Royal College of Radiologists, England).
Professor and Director of Oral & Maxillofacial Radiology
Dept of General Dentistry
Creighton University Dental School
2802 Webster St, Boyne 210
Omaha
Nebraska 68178

Tel: (402)280 5025
Fax: (402)280 5094

Julian Cooper

Re: Problem with "Orthogonal Views"

Dear Douglas,

I suspect this may relate to the ordering of the slices in your stack. I
deal mainly with CT images of the body and extremities where the top (head
end) of the scan is normally slice number 1. Orthogonal Views looks correct
for these stacks (e.g. coronal recon of a tibia appears the "right way up"
below the axial window and the sagittal recon appears to the right of the
axial window with proximal end towards the axial window and the distal
extending away to the right). This appears correct for these stacks because
ImageJ "sees" slice 1 as at the "top". This is consistent with the behaviour
of Stacks>Reslice where slice 1 will be at the top of the resulting
reconstruction.

As I understand it head scans (particularly brain) are numbered
conventionally from inferior to superior, so that slice 1 is the anatomical
bottom of the scanned volume. The left-right convention is the same as
elsewhere so that anatomical right is on the left side of the image. If your
images are numbered like this it would explain the appearances you see.

If you reverse your stack before using Orthogonal Views, the resulting
recons should be consistent, although the sagittal will still be on its
"back". It should be noted that some plugins assume slice 1 at the top and
others slice 1 at the bottom. I think that Volume Viewer should show your
recons correctly as I have to reverse extremity scans before using it.

Kind regards,

Julian

> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On
> Behalf Of Benn, Douglas K.
> Sent: 01 May 2010 16:28
> To: [hidden email]
> Subject: Problem with "Orthogonal Views"
>
>
> I wonder if anybody has the same problem with "Orthogonal
> Views"? When I do the following
>
> run("Image Sequence...", "open=[C:\\CBCT
> research\\04\\ID001122012.dcm] number=525 starting=1
> increment=1 scale=100 file=[] or=[] sort");
> selectWindow("04");
> run("Orthogonal Views");
>
> It displays 3 windows - top left is axial slices, top right
> is saggital and bottom left coronal. However, the saggital
> images are flipped 90 degrees clockwise and the coronal 180 degrees.
>
> I can correct the orientation by using
>
> selectWindow("YZ 256");
> run("Rotate 90 Degrees Left");
> selectWindow("XZ 256");
> run("Rotate... ", "angle=180 grid=1 interpolation=Bilinear");
>
> However, as soon as I select the axial window to drag the
> planes cursor, the other windows flip again.
>
> Does anybody know how to correct this?
>
> Thank you.
>
> Douglas
>
> Dr Douglas K Benn, BDS,DDS, M.Phil., Ph.D., Dipl. Dental
> Radiology (Royal College of Radiologists, England).
> Professor and Director of Oral & Maxillofacial Radiology
> Dept of General Dentistry
> Creighton University Dental School
> 2802 Webster St, Boyne 210
> Omaha
> Nebraska 68178
>
> Tel: (402)280 5025
> Fax: (402)280 5094
>

Douglas Benn-2

Re: Problem with "Orthogonal Views"

Dear Julian,

Spot on. Head slices are usually inf to superior starting from 1. I reversed my stack and as you predicted my coronal slices are correct but the saggital is still "wrong". Volume viewer does have the correct orientations. Perhaps I should contact the author of Orthogonal Views for their advice.

Thanks for your help.

Best wishes.

Douglas

Dr Douglas K Benn, BDS,DDS, M.Phil., Ph.D., Dipl. Dental Radiology (Royal College of Radiologists, England).
Professor and Director of Oral & Maxillofacial Radiology
Dept of General Dentistry
Creighton University Dental School
2802 Webster St, Boyne 210
Omaha
Nebraska 68178

Tel: (402)280 5025
Fax: (402)280 5094

-----Original Message-----
From: ImageJ Interest Group on behalf of Julian Cooper
Sent: Sat 5/1/2010 12:20 PM
To: [hidden email]
Subject: Re: Problem with "Orthogonal Views"

Dear Douglas,

I suspect this may relate to the ordering of the slices in your stack. I
deal mainly with CT images of the body and extremities where the top (head
end) of the scan is normally slice number 1. Orthogonal Views looks correct
for these stacks (e.g. coronal recon of a tibia appears the "right way up"
below the axial window and the sagittal recon appears to the right of the
axial window with proximal end towards the axial window and the distal
extending away to the right). This appears correct for these stacks because
ImageJ "sees" slice 1 as at the "top". This is consistent with the behaviour
of Stacks>Reslice where slice 1 will be at the top of the resulting
reconstruction.

As I understand it head scans (particularly brain) are numbered
conventionally from inferior to superior, so that slice 1 is the anatomical
bottom of the scanned volume. The left-right convention is the same as
elsewhere so that anatomical right is on the left side of the image. If your
images are numbered like this it would explain the appearances you see.

If you reverse your stack before using Orthogonal Views, the resulting
recons should be consistent, although the sagittal will still be on its
"back". It should be noted that some plugins assume slice 1 at the top and
others slice 1 at the bottom. I think that Volume Viewer should show your
recons correctly as I have to reverse extremity scans before using it.

Kind regards,

Julian

Rasband, Wayne (NIH/NIMH) [E]

Re: Problem with "Orthogonal Views"

In reply to this post by Douglas Benn-2

On May 1, 2010, at 11:28 AM, Benn, Douglas K. wrote:

> I wonder if anybody has the same problem with "Orthogonal Views"? When I do the following
>
> run("Image Sequence...", "open=[C:\\CBCT research\\04\\ID001122012.dcm] number=525 starting=1 increment=1 scale=100 file=[] or=[] sort");
> selectWindow("04");
> run("Orthogonal Views");
>
> It displays 3 windows - top left is axial slices, top right is saggital and bottom left coronal. However, the saggital images are flipped 90 degrees clockwise and the coronal 180 degrees.
>
> I can correct the orientation by using
>
> selectWindow("YZ 256");
> run("Rotate 90 Degrees Left");
> selectWindow("XZ 256");
> run("Rotate... ", "angle=180 grid=1 interpolation=Bilinear");
>
> However, as soon as I select the axial window to drag the planes cursor, the other windows flip again.
>
> Does anybody know how to correct this?

You can correct this in the 1.44a daily build by checking "Rotate YZ" and "Flip XZ" in the Edit>Options>DICOM dialog box.

-wayne