Representing and describing radiology images using OBI

[Shahim Essaid]

Hi all,

I am new to this group and I am not sure if I am posting to the
correct OBI mailing list. Does OBI have entities to support
describing images? I would like to create entities to describe
ultraound (and possibly other image types) image findings.

I am working on an ontology of GI endoscopy and one part of this
ontology will have to deal with the description of images. I looked
through the OBI hierarchy but I couldn't recognize the best place
where I can extend it for my use case. For example, the term
"hyperechoic focus" is referring to a thing on an image which in turn
is (possibly) referring to something real. My initial thought is that
the image things are generic dependent continuants and they could be
abstracted as "image artifact". Various artifacts can then be defined
using various qualities. Also, I thought that an "image artifact" can
be categorized as a "symbol" so that other symbols can be placed near
each other.

Looking at OBI, "symbol" has more of a word/text meaning rather than
the more general meaning. Also, "image" is an information content
entity but I couldn't find how an image can be described using OBI.

What is the best way to use or extend OBI so that I can describe the
things seen on imaging results? Or is OBI not the right ontology to
use in this case? I was thinking that an ultrasound procedure is an
investigation and the result is an image that can be described using
OBI.

Thanks,
Shahim

[Chris Stoeckert]

Dear Shahim,
Development of OBI has moved away from the branches and is focusing on
getting the first official release out so there probably won't be any
traffic on this list for the near future. DENRIE is meant to cover
information entities specific to investigations. More general entities
are being addressed by the information artifact ontology (IAO) that
might be a better place for you to start.
http://code.google.com/p/information-artifact-ontology/

Cheers.
Chris

[Bjoern Peters]

Dear Shahim,

I want to second Chris' email, and expand a bit:

A high level term like 'image' or 'image artifact' will reside in IAO.

OBI will subclass under this term, and is likely the right place for all endoscopy related terms.

Development between IAO and OBI is proceeding in parallel, and I would encourge you to subscribe to the obi-dev list. If you already have a list of terms you want to deal with, creating a tracker item would also be good.

As Chris said, we are about to finalize our 1.0 paper (it is a matter of days now), and after that is a good time to tackle new areas such as yours.

Best,

Bjoern

--
Bjoern Peters
Assistant Member
La Jolla Institute for Allergy and Immunology
9420 Athena Circle
La Jolla, CA 92037, USA
Tel: 858/752-6914
Fax: 858/752-6987
http://www.liai.org/pages/faculty-peters

[Shahim Essaid]

Dear Bjoern and Chris,

Thank you for your answers. Explaining the relationship between OBI and
IAO was helpful and I will follow up with the IAO project.

Best regards,
Shahim