Selective automated indexing of findings and diagnoses in radiology reports

William (Bill) Hersh, Mark Mailhot, Catherine Arnott-Smith, Henry Lowe

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

The recent improvements in capabilities of desktop computers and communications networks give impetus for the development of clinical image repositories that can be used for patient care and medical education. A challenge in the use of these systems is the accurate indexing of images for retrieval performance acceptable to users. This paper describes a series of experiments aiming to adapt the SAPHIRE system, which matches text to concepts in the UMLS Metathesaurus, for the automated indexing of image reports. A series of enhancements to the baseline system resulted in a recall of 63% but a precision of only 30% in detecting concepts. At this level of performance, such a system might be problematic for users in a purely automated indexing environment. However, if the ability to retrieve images in repositories based on content in their reports is desired by clinical users, and no other current systems offer this functionality, then follow-up research questions include whether these imperfect results would be useful in a completely or partially automated indexing environment and/or whether other approaches can improve upon them.

Original languageEnglish (US)
Pages (from-to)262-273
Number of pages12
JournalJournal of Biomedical Informatics
Volume34
Issue number4
DOIs
StatePublished - 2001

Fingerprint

Unified Medical Language System
Medical education
Radiology
Computer networks
Personal computers
Telecommunication networks
Computer Communication Networks
Medical Education
Patient Care
Experiments
Research

Keywords

  • Automated indexing
  • Evaluation
  • Metathesaurus
  • Natural language processing
  • SAPHIRE
  • Unified medical language system (UMLS)

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

Selective automated indexing of findings and diagnoses in radiology reports. / Hersh, William (Bill); Mailhot, Mark; Arnott-Smith, Catherine; Lowe, Henry.

In: Journal of Biomedical Informatics, Vol. 34, No. 4, 2001, p. 262-273.

Research output: Contribution to journalArticle

Hersh, William (Bill) ; Mailhot, Mark ; Arnott-Smith, Catherine ; Lowe, Henry. / Selective automated indexing of findings and diagnoses in radiology reports. In: Journal of Biomedical Informatics. 2001 ; Vol. 34, No. 4. pp. 262-273.
@article{cab12d8605af42118b5715b4b4a9d394,
title = "Selective automated indexing of findings and diagnoses in radiology reports",
abstract = "The recent improvements in capabilities of desktop computers and communications networks give impetus for the development of clinical image repositories that can be used for patient care and medical education. A challenge in the use of these systems is the accurate indexing of images for retrieval performance acceptable to users. This paper describes a series of experiments aiming to adapt the SAPHIRE system, which matches text to concepts in the UMLS Metathesaurus, for the automated indexing of image reports. A series of enhancements to the baseline system resulted in a recall of 63{\%} but a precision of only 30{\%} in detecting concepts. At this level of performance, such a system might be problematic for users in a purely automated indexing environment. However, if the ability to retrieve images in repositories based on content in their reports is desired by clinical users, and no other current systems offer this functionality, then follow-up research questions include whether these imperfect results would be useful in a completely or partially automated indexing environment and/or whether other approaches can improve upon them.",
keywords = "Automated indexing, Evaluation, Metathesaurus, Natural language processing, SAPHIRE, Unified medical language system (UMLS)",
author = "Hersh, {William (Bill)} and Mark Mailhot and Catherine Arnott-Smith and Henry Lowe",
year = "2001",
doi = "10.1006/jbin.2001.1025",
language = "English (US)",
volume = "34",
pages = "262--273",
journal = "Journal of Biomedical Informatics",
issn = "1532-0464",
publisher = "Academic Press Inc.",
number = "4",

}

TY - JOUR

T1 - Selective automated indexing of findings and diagnoses in radiology reports

AU - Hersh, William (Bill)

AU - Mailhot, Mark

AU - Arnott-Smith, Catherine

AU - Lowe, Henry

PY - 2001

Y1 - 2001

N2 - The recent improvements in capabilities of desktop computers and communications networks give impetus for the development of clinical image repositories that can be used for patient care and medical education. A challenge in the use of these systems is the accurate indexing of images for retrieval performance acceptable to users. This paper describes a series of experiments aiming to adapt the SAPHIRE system, which matches text to concepts in the UMLS Metathesaurus, for the automated indexing of image reports. A series of enhancements to the baseline system resulted in a recall of 63% but a precision of only 30% in detecting concepts. At this level of performance, such a system might be problematic for users in a purely automated indexing environment. However, if the ability to retrieve images in repositories based on content in their reports is desired by clinical users, and no other current systems offer this functionality, then follow-up research questions include whether these imperfect results would be useful in a completely or partially automated indexing environment and/or whether other approaches can improve upon them.

AB - The recent improvements in capabilities of desktop computers and communications networks give impetus for the development of clinical image repositories that can be used for patient care and medical education. A challenge in the use of these systems is the accurate indexing of images for retrieval performance acceptable to users. This paper describes a series of experiments aiming to adapt the SAPHIRE system, which matches text to concepts in the UMLS Metathesaurus, for the automated indexing of image reports. A series of enhancements to the baseline system resulted in a recall of 63% but a precision of only 30% in detecting concepts. At this level of performance, such a system might be problematic for users in a purely automated indexing environment. However, if the ability to retrieve images in repositories based on content in their reports is desired by clinical users, and no other current systems offer this functionality, then follow-up research questions include whether these imperfect results would be useful in a completely or partially automated indexing environment and/or whether other approaches can improve upon them.

KW - Automated indexing

KW - Evaluation

KW - Metathesaurus

KW - Natural language processing

KW - SAPHIRE

KW - Unified medical language system (UMLS)

UR - http://www.scopus.com/inward/record.url?scp=0035564683&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035564683&partnerID=8YFLogxK

U2 - 10.1006/jbin.2001.1025

DO - 10.1006/jbin.2001.1025

M3 - Article

C2 - 11977808

AN - SCOPUS:0035564683

VL - 34

SP - 262

EP - 273

JO - Journal of Biomedical Informatics

JF - Journal of Biomedical Informatics

SN - 1532-0464

IS - 4

ER -