[Dcmlib] Fwd: Improving IPPSorter

[Bartosz Wiklak]

---------- Forwarded message ----------
From: Bartosz Wiklak <bwiklak at gmail.com>
Date: Thu, Nov 27, 2008 at 9:09 AM
Subject: Improving IPPSorter
To: dcmlib at creatis.insa-lyon.fr, mathieu.malaterre at gmail.com


HI,

I'm thinking of improving IPPSorter but first I would like to know
what you think about it.
First of all, IPPSorter should take under consideration if all slices
provided by user have the same Frame of Reference UID (0020,0052), if
not, their (IOP, IPP) pairs should not be compared.

>From DICOM PS 3.3 - 2007, Page 287:

"
C.7.4.1.1.1 Frame Of Reference UID
The Frame of Reference UID (0020,0052) shall be used to uniquely
identify a frame of reference
for a series. Each series shall have a single Frame of Reference UID.
However, multiple Series
within a Study may share a Frame of Reference UID. All images in a
Series that share the same
Frame of Reference UID shall be spatially related to each other.
Notes: 1. Previous versions of this Standard defined a Data Element
"Location", which has been
retired. Frame of Reference UID provides a completely unambiguous
identification of the image
location reference used to indicate position.
2. A common Frame of Reference UID may be used to spatially relate
localizer images with a
set of axial images. However, in some cases (eg. multiple localizer
images being related to a
single set of axial images) a common Frame of Reference UID may not be
sufficient. The
Referenced Image Sequence (0008,1140) provides an unambiguous method
for relating
localizer images.
"


If not all slices have the same FrameOfReferenceUID, IPPSorter should
fail with some kind of error indicating, that data is not generally
invalid, but we have N number of different volumes.
It would be grate if user could get some kind of list of f.e.
FrameOfReferenceUID's and choose one of them then sort again.

Also sometimes I have MR series in which tho kind of expositions (
f.e. Echo time 11 and echo tme 88 ) are placed one slice after another
i.e.:
slice1 TE 11
slice1 TE 88
slice2 TE 11
...


IPPSorter should not fail in this situation but provide user chance to
select data, he wants to visualize.

Have you chosen some kind of  error system to provide user errors and
their descriptions after method execution?

Bartosz Wiklak