About CS&CE
Prospective Students
Research
Staff
Dept Comp Sci & Comp Eng
Contact Details
La Trobe University
Victoria 3086
AUSTRALIA
Tel: +61 3 9479 1107
Fax: +61 3 9479 3060
Email: info
@cs.latrobe.edu.au
|
 |
Grants - Detail
Department of Computer Science & Computer Engineering
| Medical Information Systems Interoperability and Retrieval |
The integration of different documents and resources in health information systems into a standardised knowledge representation has created a key area in health informatics e-research in the last few years. The establishment of an ontology in medical domains such as the UMLS (Unified Medical Language Systems) forms the conceptual standard that can be used to support data and knowledge interoperability and sharing.
In this project, we will address three major issues in the development of integrated medical information systems:
(i) The establishment of techniques and representations that allows the integration of both information as well as semantic properties of medical information. We will use UMLS Semantic Network as our base ontology in this project and perform metadata annotations of medical documents such as the Therapeutic Guidelines, Australian Medicines Handbook, and the National Prescribing Service into this underlying ontology.
(ii) The need to have an efficient retrieval technique for the integrated information. This is especially important since the size of the integrated information will likely to grow very large, and the client side that needs to access the information may be limited in hardware resources (eg. a palmtop). We will adopt our algorithms that have been developed using VPAC facilities in our earlier VPAC expertise project to derive Materialised Ontology Views (MOVE) from the integrated information. These views, which can be seen as optimised subsets of information that are generated based on a number of requirement specific optimisation schemes, will form the basis for medical information retrieval.
(iii) The need to for an effective means of displaying integrated medical information to the user, which enables efficient navigation and data viewing. We will employ dynamic indexing techniques which dynamically cluster and index the texts within medical documents based on the underlying (sub) ontology.
It is expected that at the end of this project we will be able to prototype the integration of some of the Australian medical sources with an underlying standard conceptual structure. We will apply and validate our ontology derivation optimisation schemes and propose a new optimisation scheme that suits the medical domain. Finally, we will prototype the visualisation of the materialised views using a dynamic indexing technique for navigation.
|
|
|
