Automatic semantic labeling of medical texts with feature structures

Authors:
Agnieszka Mykowiecka;Małgorzata Marciniak
Affiliations:
Institute of Computer Science, Polish Academy of Sciences, Warsaw, Poland;Institute of Computer Science, Polish Academy of Sciences, Warsaw, Poland
Venue:
TSD'11 Proceedings of the 14th international conference on Text, speech and dialogue
Year:
2011

Citing 6
Cited 1

Automating Accreditation of Medical Web Content

Proceedings of the 2008 conference on ECAI 2008: 18th European Conference on Artificial Intelligence
Semantic Annotation of City Transportation Information Dialogues Using CRF Method

TSD '09 Proceedings of the 12th International Conference on Text, Speech and Dialogue
Rule-based information extraction from patients' clinical data

Journal of Biomedical Informatics
Building a semantically annotated corpus of clinical texts

Journal of Biomedical Informatics
Corpus design for biomedical natural language processing

ISMB '05 Proceedings of the ACL-ISMB Workshop on Linking Biological Literature, Ontologies and Databases: Mining Biological Semantics
Domain Model for Medical Information Extraction--The LightMedOnt Ontology

Aspects of Natural Language Processing

Towards morphologically annotated corpus of hospital discharge reports in Polish

BioNLP '11 Proceedings of BioNLP 2011 Workshop

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents the results of testing two approaches in the automatic semantic labeling of medical data. For a chosen domain (diabetic patients' discharge records) a set of domain related concepts was identified. The annotated resource is the result of a rule based application, that relies on the results of two related rule based information extraction (IE) systems, post processed in a way that makes the label structures simpler, and the boundaries of annotations more precise. The second application is a machine learning (CRF) approach in which the results of the first application are used as training data. Both applications were evaluated by comparing to manually corrected documents.