Related Experiment Videos
Understanding of medico-technical reports.
1Laboratoire de Biomathématique, Faculté de Médecine, Université de la Méditerranée, 27 Bd. Jean Moulin, 1385, Marseille, France. mtcd@medecine.univ-mrs.fr
Artificial Intelligence in Medicine
|January 29, 2000
Summary
The Aristotle project develops an automated system to process medical text, converting it into a structured semantic format for research. This system overcomes data processing challenges, enabling better utilization of clinical documentation.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Knowledge Representation
Background:
- Vast amounts of unstructured text are generated in medical practice.
- Underutilization of this valuable data for research stems from significant information processing challenges.
Purpose of the Study:
- To develop an automated data system, the Aristotle project, for creating semantic representations of medical text.
- To enable the extraction and structuring of information from clinical documentation for research purposes.
Main Methods:
- Implementing a syntactic-semantic interpreter to process text sentence by sentence.
- Utilizing a knowledge base for semantic interpretation and object-relation identification.
- Employing an Assembler module to integrate sentence-level meanings into a coherent output structure.
Main Results:
- The system successfully identifies objects, their properties, and relationships within medical text.
- A canonical semantic representation of the text is generated, facilitating data analysis.
- The syntactic-semantic interpreter effectively links lexical units and resolves meaning.
Conclusions:
- The Aristotle project demonstrates a viable approach to automatically processing and structuring medical text.
- This semantic representation system enhances the potential for leveraging clinical text data in research.
- Automated semantic analysis offers a solution to the under-usage of medical documentation in research.