Using The Semantic Web and Proof Technologies to Reduce Errors in Radiological Procedure Orders

from http://www.scarnet.net/trip/pdf/TRIP_Abstracts.pdf

Using The Semantic Web and Proof Technologies to
Reduce Errors in Radiological Procedure Orders

John S. Devitt, PhD, Agfa Healthcare
Helen H. Chen, PhD; Jos De Roo, MSc

Topic: Databases and Integration

Introduction:
The obvious benefits of radiological procedures often
come with harmful side effects and high costs. The
Radiation Protection 118 - Referral Guideline for
Imaging [1], cautions against the wasteful usage
of radiology such as unnecessary or too frequently
occurring investigations, investigations that result
in little or no impact on patient management, and
wrong investigations. In each case, the patient may be
subjected unnecessarily to irradiation and scarce medical
resources are wasted.
This paper presents a highly scalable framework based
on semantic web technologies for integrating the
medical knowledge contained in such guidelines and
heterogeneous patient information. It demonstrates how
proof technology can be used to significantly reduce
the wasteful or wrong usage of radiological procedures
by detecting inconsistencies. The techniques involve
integrating knowledge across a wide range of sources.

Methods:
Our study uses N3 notation [2] and the OWL Web
Ontology Language [3] to transform the Referral
Guideline for Imaging [1] into a machine-readable
medical knowledge base based on ontologies. The
ontology (radGuideline) associates clinical problems
with preferred investigations, and captures in rules
the recommendations and comments that contain the
imperative wisdom of the medical community. The
probabilistic nature of the data is captured through the
use of a special vocabulary developed to capture the
explicit meaning of the recommendation. The Ontology
enables the reuse of existing medical knowledge by
linking this specialized information with other medical
ontologies such as SNOMED CT [4] for standard
description for clinical problems and the patient's
history.
Medical histories are often stored in various databases
in a hospital, and even at different locations in a
community healthcare setting. The ontology provides
structure and meaning of patient medical data and
enables the proof engines to understand and utilize
patient data without having to deal with the differences
in data format or location.
Semantic web engines Euler [5] and CWM [6] are able
to generate proofs based on the given patient conditions
(medical history and physical condition) and the
radGuideline ontology.

Results:
We apply these techniques to a number of clinical
scenarios pertaining to radiological orders for female
patients at reproductive age. The system is able to
validate the proposed order against the specific patient
demographic data, medical record that are stored at
dispersed locations and rules prescribed by the Radiation
Referral Guideline.

Discussion:
The true potential of the proposed method is its capability
of integrating cross-domain knowledge and data
seamlessly on explicit and unambiguous terms expressed
in ontologies. The explanations generated by proof
engines provide evidence to clinicians for a decision. It
can even provide alternatives. The final decision still
lies in the hands of a clinician, but making such key
information and evidence readily available to him/her is
extremely important in the presence of the large volume
of data. It can help to eliminate medical errors caused by
overlooking vital facts when making medical decisions.

Conclusion:
By using semantic web technology, we are able to link
different medical knowledge-bases to enlarge our medical
expertise in order to deliver relevant information to the
clinician, when and where they need it. The information
system is also capable of raising flags when a prescribed
radiological procedure is not safe for a patient or is not
normally the most effective means according to the latest
medical advancement or protocol.


References:

[1] RADIATION PROTECTION Referral guidelines for imaging
 
http://europa.eu.int/comm/energy/nuclear/radioprotection/publication/doc/118_en.pdf

[2] Semantic Web Tutorial Using N3
     http://www.w3.org/2000/10/swap/doc/

[3] OWL Web Ontology Language Overview
     http://www.w3.org/TR/owl-features/

[4] SNOMED CT
     http://www.snomed.org/snomedct/

[5] Euler proof mechanism
     http://www.agfa.com/w3c/euler/

[6] Cwm
    http://www.w3.org/2000/10/swap/doc/cwm.html

-- 
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/

Received on Tuesday, 1 March 2005 16:02:42 UTC