Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dr. Nadkarni replies

Dr. Nadkarni replies 512 Letters to the Editor Letters to the Editor JAMIA We used automated term composition and the UMLS UMLS Concept Indexing for Production to assess match rates. In addition, we looked at match Databases: A Feasibility Study rates on our total 14,044 terms based on filtering using the UMLS semantic types. Comparison of the data from the two studies To the Editor:—In the recently published study by (Table 1) reveals striking similarities. Nadkarni et al., the authors used text-mining soft- ware to extract concepts from clinical documents. What we recognized in 1999, which was omitted Matching of these concepts was attempted with the from the analysis of Nadkarni et al., was that other UMLS 99 Metathesaurus. Matches were then catego- metrics are important in the clinical interpretation of rized as true positives (TP), false positives (FP), true these data. Representing the data as shown in negatives (TN), and false negatives (FN) from 8,745 Table 1 allows for useful combinations. The true- terms in a “training set” and 1,701 terms in a “test positive rate is the number of true positives divided set,” for a total of 10,446 terms. True positives were by the sum of true positives and false http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of the American Medical Informatics Association Oxford University Press

Loading next page...
 
/lp/oxford-university-press/dr-nadkarni-replies-X4JlwXpFKq

References (3)

Publisher
Oxford University Press
Copyright
American Medical Informatics Association
ISSN
1067-5027
eISSN
1527-974X
DOI
10.1136/jamia.2001.0080514
Publisher site
See Article on Publisher Site

Abstract

512 Letters to the Editor Letters to the Editor JAMIA We used automated term composition and the UMLS UMLS Concept Indexing for Production to assess match rates. In addition, we looked at match Databases: A Feasibility Study rates on our total 14,044 terms based on filtering using the UMLS semantic types. Comparison of the data from the two studies To the Editor:—In the recently published study by (Table 1) reveals striking similarities. Nadkarni et al., the authors used text-mining soft- ware to extract concepts from clinical documents. What we recognized in 1999, which was omitted Matching of these concepts was attempted with the from the analysis of Nadkarni et al., was that other UMLS 99 Metathesaurus. Matches were then catego- metrics are important in the clinical interpretation of rized as true positives (TP), false positives (FP), true these data. Representing the data as shown in negatives (TN), and false negatives (FN) from 8,745 Table 1 allows for useful combinations. The true- terms in a “training set” and 1,701 terms in a “test positive rate is the number of true positives divided set,” for a total of 10,446 terms. True positives were by the sum of true positives and false

Journal

Journal of the American Medical Informatics AssociationOxford University Press

Published: Sep 1, 2001

There are no references for this article.