Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Regarding: "DICOM Structured Reporting - What's it good for?"

20 views
Skip to first unread message

bwi...@gmail.com

unread,
Mar 9, 2009, 9:58:05 AM3/9/09
to
Hi,

My personal DICOM guru, David Clunie, points out many advantages of
using DICOM-SR as universal report format.
It is true, that in a complicated study it is useful format and also,
f.e. in USG it can be generated semiautomatic.

I've recently read very interesting thread
http://groups.google.com/group/alt.image.medical/browse_thread/thread/df26ddccfa7eb917?tvc=2&q=SR#

I'm writing dicom viewer in a small company. As for now, our software
was able to store reports as rtf documents (relationship with images
was stored in database) or as a pdf embedded into dicom.

As I stated before, I'm very influenced by David Clunie's ideas and
his point of view and I see many disadvantages of our previous
attitude.
Regarding thread "DICOM Structured Reporting - What's it good for?",
our viewer is supposed to be a universal dicom viewer, and we don't
want to focus on any particular type of study. We would also like to
provide as simple as possible method to write a basic report. We don't
want to enforce physician to fill complex forms.

How do you think, what form should BasicTextSr document have to be as
simple as possible and also, to still be a valid and reasonable
structured report?

What can be done to give physician ability to write text in non
English character set and still leave other vendor's viewers
opportunity to present such report correctly ( f.e. in countries
having nonlatin characters such as Japan, Russia or China, vendors are
forced to implement appropriate character sets, but in countries like
Poland, in witch only fey characters are not from standard ASCII,
there is no such implementations especially from big vendors ).

Do you have any ideas?

Thanks in advance
Bartosz Wiklak

bwi...@gmail.com

unread,
Mar 12, 2009, 7:51:41 AM3/12/09
to
On 9 Mar, 14:58, bwik...@gmail.com wrote:
> Hi,
>
> My personal DICOM guru, David Clunie, points out many advantages of
> using DICOM-SR as universal report format.
> It is true, that in a complicated study it is useful format and also,
> f.e. in USG it can be generated semiautomatic.
>
> I've recently read very interesting threadhttp://groups.google.com/group/alt.image.medical/browse_thread/thread...

>
> I'm writing dicom viewer in a small company. As for now, our software
> was able to store reports as rtf documents (relationship with images
> was stored in database) or as a pdf embedded into dicom.
>
> As I stated before, I'm very influenced by David Clunie's ideas and
> his point of view and I see many disadvantages of our previous
> attitude.
> Regarding thread "DICOM Structured Reporting - What's it good for?",
> our viewer is supposed to be a universal dicom viewer, and we don't
> want to focus on any particular type of study. We would also like to
> provide as simple as possible method to write a basic report. We don't
> want to enforce physician to fill complex forms.
>
> How do you think, what form should BasicTextSr document have to be as
> simple as possible and also, to still be a valid and reasonable
> structured report?
>
> What can be done to give physician ability to write text in non
> English character set and still leave other vendor's viewers
> opportunity to present such report correctly ( f.e. in countries
> having nonlatin characters such as Japan, Russia or China, vendors are
> forced to implement appropriate character sets, but in countries like
> Poland, in witch only fey characters are not from standard ASCII,
> there is no such implementations especially from big vendors ).
>
> Do you have any ideas?
>
> Thanks in advance
> Bartosz Wiklak

I thought this post will start interesting discussion about
implementation of SR.
Does anyone have something to say about the problem of complexity vs.
reporting speed?

Maybe such thread was posted before, could you point me the right
directions?

Harry Solomon

unread,
Apr 22, 2009, 12:55:10 AM4/22/09
to
Hello, Bartosz -

I will respond to your character set question.

> > What can be done to give physician ability to write text in non
> > English character set and still leave other vendor's viewers
> > opportunity to present such report correctly ( f.e. in countries
> > having nonlatin characters such as Japan, Russia or China, vendors are
> > forced to implement appropriate character sets, but in countries like
> > Poland, in witch only fey characters are not from standard ASCII,
> > there is no such implementations especially from big vendors ).

The DICOM Standard clearly describes how to use ISO IR 101 / ISO
8859-2 / Latin-2 to encode Polish and other central European
languages. But you are correct that this has generally not been
implemented, although many vendors do support ISO IR 100 / ISO
8859-1 / Latin-1 as an extended character set. At this point in time,
it is unlikely that 8859-2, or indeed any additional 8859 series
character sets, will be implemented in products, as pretty much all
work is now focused on transitioning implementations to Unicode as a
general approach for all languages. (Note that ISO itself has dropped
all further work on 8859, and is working only on Unicode.)

- Harry Solomon

0 new messages