Related Experiment Videos
A multi-lingual architecture for building a normalised conceptual representation from medical language
P Zweigenbaum1, B Bachimont, J Bouaud
1DIAM--INSERM U.194, Hôpitaux de Paris.
Proceedings. Symposium on Computer Applications in Medical Care
|January 1, 1995
Summary
The MENELAS project developed a prototype for analyzing medical texts in patient discharge summaries (PDSs). It aims to improve information access through normalized, domain-independent, and language-independent representations.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Computational Linguistics
Background:
- Patient discharge summaries (PDSs) contain valuable clinical information.
- Accessing and utilizing information from unstructured PDSs is challenging.
- Current methods for PDS analysis are often domain-specific and language-dependent.
Purpose of the Study:
- To design and implement a prototype system (MENELAS) for analyzing natural language PDSs.
- To enhance access to information within medical texts.
- To explore the practical application of key principles in medical text analysis.
Main Methods:
- Developing a prototype for natural language analysis of medical texts.
- Implementing a normalized conceptual representation for medical information.
- Focusing on domain-independent language analysis and language-independent conceptual representation.
Main Results:
- Demonstrated the feasibility of creating a normalized conceptual representation from PDSs.
- Achieved progress in developing domain-independent natural language analysis techniques.
- Identified practical challenges and solutions in implementing the project's core principles.
Conclusions:
- The MENELAS project successfully demonstrated a prototype for analyzing patient discharge summaries.
- Normalized, domain-independent, and language-independent representations are key to maximizing the utility and reusability of medical text analysis.
- Further research is needed to address implementation challenges and refine the system.