BISC Seminar Announcement, December 2nd, 1999, 4-5pm, 310 Soda

Frank Hoffmann (fhoffman@cs.berkeley.edu)
Thu, 25 Nov 1999 03:14:43 +0100 (MET)

*********************************************************************
Berkeley Initiative in Soft Computing (BISC)
*********************************************************************
************************************************************
B I S C S e m i n a r A n n o u n c e m e n t
************************************************************


Integrating Knowledge and Handling Imprecision in Medical and
Satellite Image Analysis

Speaker :

Lawrence O. Hall

Visiting UC Berkeley
Dept. of Computer Science and Engineering, ENB 118
University of South Florida
E-mail: hall@csee.usf.edu

Date: Thursday, December 2nd, 1999
Time: 4-5pm
Location : 320 Soda Hall

Abstract

Physicians are very good at quickly recognizing abnormalities in
medical images. They may "see" small tumor's in brain images. People
analyzing satellite images may immediately recognize a specific region
as a phytoplankton bloom or a particular type of growth on land or a set of
vehicles despite differences in the quality of the images. It is easy
for people to recognize objects in camera images of the same region
taken over time. Our work focuses on some ways in which knowledge
and approximate reasoning may be utilized in recognizing expected
objects of interest in images taken of the same area over time.

For example, finding abnormalities in medical images of the human
body can be made easier by including domain knowledge. Spatial
knowledge and expected findings may be exploited to a
degree. Methods of quickly grouping data and labeling the homogeneous
groups or clusters can be utilized to remove "expected" data.

The handling of imprecise boundaries
is done by using fuzzy approaches. The amount of overlap of clusters
is a clue to how well the data has been divided into unique groups.
Domain knowledge about how clustering or learning algorithms work on a
set of training images can be incorporated as rule based knowledge to
guide processing. Data may be analyzed in a hierarchical manner with
successive refinement of knowledge guided segmentation. Different
knowledge may be applied at each level of the hierarchy with some of
the knowledge applying to feature space as affected by an algorithm
and some applying to parameter space.

Experimental results from the domain of magnetic resonance
images of the human brain are promising in automatically measuring
brain tumor volume over time from gadolinium enhanced T1, T2 and
proton density weighted images. Results from color satellite images show that
red tide and green river may be tracked over time on the waters
off the West Florida shelf.

**********************************************************************
Please direct questions with regard to the contents of the talk
and request for papers to the speaker.
**********************************************************************

-- 
---------------------------------------------------------------------------
Frank Hoffmann                               UC Berkeley
Computer Science Division                    Department of EECS
Email: fhoffman@cs.berkeley.edu              phone: 1-510-642-8282
URL: http://http.cs.berkeley.edu/~fhoffman   fax:  1-510-642-5775
---------------------------------------------------------------------------
If you ever want to remove yourself from this mailing list,
you can send mail to <Majordomo@EECS.Berkeley.EDU> with the following
command in the body of your email message:
unsubscribe bisc-group
or from another account,
unsubscribe bisc-group <your_email_adress>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Do NOT send unsubscribe requests to bisc-group@cs.berkeley.edu
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

############################################################################ This message was posted through the fuzzy mailing list. (1) To subscribe to this mailing list, send a message body of "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at (2) To unsubscribe from this mailing list, send a message body of "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com" to listproc@dbai.tuwien.ac.at (3) To reach the human who maintains the list, send mail to fuzzy-owner@dbai.tuwien.ac.at (4) WWW access and other information on Fuzzy Sets and Logic see http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html