Automatic identification of heart failure diagnostic criteria, using text analysis of clinical notes from electronic health records

Automatic identification of heart failure diagnostic criteria, using text analysis of clinical notes from electronic health records

Highlights ► Framingham heart failure criteria can be reliably detected in EHR clinical notes. ► Entity extractors, expert annotations, and guidelines are created simultaneously. ► We mitigate initial low recall of human coders when creating reference annotations.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of medical informatics (Shannon, Ireland) Ireland), 2014-12, Vol.83 (12), p.983-992
Hauptverfasser: Byrd, Roy J, Steinhubl, Steven R, Sun, Jimeng, Ebadollahi, Shahram, Stewart, Walter F
Format: Artikel
Sprache:eng
Schlagworte:
Automatic Data Processing
Cohort Studies
Data Mining - methods
Data Mining - utilization
Diagnostic criteria
Electronic health records
Electronic Health Records - utilization
Heart failure
Heart Failure - diagnosis
Humans
Internal Medicine
Natural Language Processing
Other
Population Surveillance
Primary Health Care
Text mining
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Highlights ► Framingham heart failure criteria can be reliably detected in EHR clinical notes. ► Entity extractors, expert annotations, and guidelines are created simultaneously. ► We mitigate initial low recall of human coders when creating reference annotations.
ISSN:1386-5056
1872-8243
DOI:10.1016/j.ijmedinf.2012.12.005