[Neuroimaging] DICOM Orientation and World<-->Image coordinate transformation

[Athanasios Anastasiou]

Hello Matthew & Steven

Thank you for your email. Of course I am missing the third column :( I am
paying too much attention on the two numbers I am after right now, to bring
the contour right where it should be when plotting it over the image.

Thank you for your help, I will have another go at establishing the matrix
with the helpful comments provided here.

All the best
AA

On 6 Sep 2016 18:25, "Matthew Brett" <matthew.brett at gmail.com> wrote:

> Hi,
>
> On Tue, Sep 6, 2016 at 6:34 AM, Steve Pieper <pieper at isomics.com> wrote:
> > Hi Athanasios -
> >
> > To get the scan direction you'll need to look at the relative
> > ImagePositionPatient points from slice to slice.  Note that the scan
> > direction is not always the cross product of the row and column
> orientations
> > since the scan may go in the other direction from a right handed cross
> > product or the slices can be sheared (or even at arbitrary locations)..
> > There are lots of other things that can happen too, like irregular
> spacing,
> > missing slices, etc, but usually just normalizing the vector between your
> > origin and any slice in the scan will be what you want.
> >
> > This code will give you an idea:
> >
> > https://github.com/Slicer/Slicer/blob/master/Modules/
> Scripted/DICOMPlugins/DICOMScalarVolumePlugin.py#L195-L216
> >
>
> From your code, you are missing a valid third column for your affine.
> I believe that column will be all zeros from your code. This is what
> the later part of the DICOM orientation page is talking about, and
> what Steve is referring to as the "slice direction".
>
> Steve is quite right that the slice direction need not be the
> cross-product of the first two, and the DICOM information can often
> tell you what that slice direction vector is, but assuming for a
> moment that it is the cross product, and that you are looking at the
> first slice of the volume, then you'd want something like:
>
> """
> import numpy as np
>
> ImageOrientationPatient = [0.999857, 0.00390641, 0.0164496,
>                            -0.00741602, 0.975738, 0.218818]
>
> ImagePositionPatient = [-127.773, -105.599, -94.5758]
>
> PixelSpacing = [0.4688, 0.4688]
>
> slice_spacing = 3.0  # ?
>
> # Make F array from DICOM orientation page
> F = np.fliplr(np.reshape(ImageOrientationPatient, (2, 3)).T)
> rotations = np.eye(3)
> rotations[:, :2] = F
> # Third direction cosine from cross-product of first two
> rotations[:, 2] = np.cross(F[:, 0], F[:, 1])
> # Add the zooms
> zooms = np.diag(PixelSpacing + [slice_spacing])
>
> # Make the affine
> affine = np.diag([0., 0, 0, 1])
> affine[:3, :3] = rotations.dot(zooms)
> affine[:3, 3] = ImagePositionPatient
>
> np.set_printoptions(precision=4, suppress=True)
> print(affine)
> """
>
> But - Steve's suggestion is more general - this code is just to give
> you an idea.
>
> Best,
>
> Matthew
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20160908/78157446/attachment.html>