Related Experiment Videos
Automated coding of patient discharge summaries using conceptual graphs
D Delamarre1, A Burgun, L P Seka
1Laboratoire Informatique Médicale, Faculté de Médecine, Université de Rennes I, France.
Methods of Information in Medicine
|September 1, 1995
Summary
This study presents an automated system for coding coronary disease patient discharge summaries into ICD-9-CM. The natural language processing system streamlines medical coding, improving efficiency and accuracy in information retrieval.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Health Information Management
Background:
- Manual encoding of patient discharge summaries is time-consuming.
- Accurate indexing and classification are crucial for medical information retrieval.
- Existing methods lack efficiency in processing complex medical documents.
Purpose of the Study:
- To develop an automated coding system for coronary disease patient discharge summaries.
- To translate medical terms into a conceptual graph model for classification.
- To comply with International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding rules.
Main Methods:
- Developed an automated coding system within the European AIM MENELAS project.
- Utilized a natural-language understanding system with conceptual graph formalism.
- Implemented a two-step processing scheme: recognition by matching and selection by coding priorities.
Main Results:
- Successfully translated classification terms into the conceptual graph model.
- Demonstrated compliance with ICD-9-CM coding rules.
- The system provides an objective evaluation for natural language understanding in medical contexts.
Conclusions:
- The automated system significantly improves the efficiency of encoding coronary disease patient discharge summaries.
- The natural language processing approach offers a robust method for medical classification.
- This system enhances objective assessment of natural language understanding in healthcare.