W3C home > Mailing lists > Public > public-rif-wg@w3.org > December 2005

[Use Case] Fuzzy Reasoning with Brain Anatomical Structures

From: Giorgos Stamou <gstam@softlab.ntua.gr>
Date: Tue, 6 Dec 2005 19:15:32 +0200
Message-Id: <200512061717.jB6HHq0A004097@theseas.softlab.ece.ntua.gr>
To: <public-rif-wg@w3.org>
The Use Case below is an example showing the need for uncertainty
representation in rules. Since uncertainty handling is not a Phase 1 or
Phase 2 characteristic of the rule language but an extensibility
requirement, we only give here a very short description of the idea of the
Use Case. Of course there is a background work behind it that could be
presented if necessary.


The work has been done in collaboration of IVML-NTUA with Christine
Golbreich and Ammar Mechouch (Laboratory in Medical Informatics, University
of Rennes). Many thanks.







** Fuzzy Reasoning with Brain Anatomical Structures


* Description


Neuroimaging applications aim at developing new methods for assisting the
labeling of the brain cortex structures in MRI (Magnetic Resonance Images).
In such applications the brain cortex can be automatically segmented into
various parts, but the problem remains to identify these parts.


 In order to enhance and assist the labeling process, knowledge of the brain
anatomy has been created in the form of ontologies. Furthermore, a rule base
has been created that further captures the dependencies between the
relations of the brain cortex structures required [1]. Other rules are
defined for validating the knowledge consistency. Unfortunately, the
numerical tools that perform image segmentation and object identification
attach to objects a truth-value, ranging from 0 to 1, regarding the
membership degree that the image objects belong to a certain concept and the
degree of asserted relations between them.


* Implications:


- The logical language should be capable of capturing and representing
partially truth knowledge.


- The reasoning engine should be capable of handling such knowledge, in
order to deduce new truth-values for the membership of objects to
predicates, based on the rules specified, or to validate the information
automatically acquired by the numerical tools.


[1] C. Golbreich, O. Bierlaire, O. Dameron, B. Gibaud, Use Case: 

Ontology with rules for identifying brain anatomical structures,






Received on Tuesday, 6 December 2005 17:18:07 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:07:18 UTC