Text mining for clinical support

In 2013, the Dahlgren Memorial Library (DML) at the Georgetown University Medical Center began using text mining software to enable its clinical informationists to quickly retrieve specific, relevant information from MEDLINE abstracts while on patient rounds. In 2013, DML licensed the use of the Lin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of the Medical Library Association 2019-10, Vol.107 (4), p.603-605
Hauptverfasser: Hartmann, Jonathan, Van Keuren, Linda
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In 2013, the Dahlgren Memorial Library (DML) at the Georgetown University Medical Center began using text mining software to enable its clinical informationists to quickly retrieve specific, relevant information from MEDLINE abstracts while on patient rounds. In 2013, DML licensed the use of the Linguamatics I2E text-mining program, and DML's clinical informationist began using it to text mine MEDLINE abstracts on patient rounds. In 2015, DML installed I2E on a server at Georgetown and negotiated with Elsevier to obtain the right to download and text mine the full text of clinical journals in ScienceDirect to support clinical decision making. In 2016, the license agreements for the and the platform were modified to allow text mining. In 2018, PubMed Central open access content was added to the Linguamatics license. DML's informationists found that they were able to quickly find useful information that was not retrievable by traditional methods, and clinicians reported the information was valuable. The ability to text mine MEDLINE abstracts and selected journal articles on patient rounds has allowed DML's clinical informationists to quickly search large amounts of medical literature that can be used to answer physicians' clinical questions. DML plans to acquire additional journal articles from selected publishers in the future, which should increase the usefulness of the project.
ISSN:1536-5050
1558-9439
DOI:10.5195/jmla.2019.758