AutoScribe: Extracting Clinically Pertinent Information from Patient-Clinician Dialogues.
Machine Learning
Medical Informatics
Medical Records
Journal
Studies in health technology and informatics
ISSN: 1879-8365
Titre abrégé: Stud Health Technol Inform
Pays: Netherlands
ID NLM: 9214582
Informations de publication
Date de publication:
21 Aug 2019
21 Aug 2019
Historique:
entrez:
24
8
2019
pubmed:
24
8
2019
medline:
5
9
2019
Statut:
ppublish
Résumé
We present AutoScribe, a system for automatically extracting pertinent medical information from dialogues between clinicians and patients. AutoScribe parses the dialogue and extracts entities such as medications and symptoms, using context to predict which entities are relevant, and automatically generates a patient note and primary diagnosis.
Identifiants
pubmed: 31438207
pii: SHTI190510
doi: 10.3233/SHTI190510
doi:
Types de publication
Journal Article
Langues
eng