A maximum likelihood approach to electronic health record phenotyping using positive and unlabeled patients

OBJECTIVE: Phenotyping patients using electronic health record (EHR) data conventionally requires labeled cases and controls. Assigning labels requires manual medical chart review and therefore is labor intensive. For some phenotypes, identifying gold-standard controls is prohibitive. We developed a...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:J Am Med Inform Assoc
Hlavní autoři: Zhang, Lingjiao, Ding, Xiruo, Ma, Yanyuan, Muthu, Naveen, Ajmal, Imran, Moore, Jason H, Herman, Daniel S, Chen, Jinbo
Médium: Artigo
Jazyk:Inglês
Vydáno: Oxford University Press 2019
Témata:
Research and Applications
On-line přístup:https://ncbi.nlm.nih.gov/pmc/articles/PMC6913222/
https://ncbi.nlm.nih.gov/pubmed/31722396
https://ncbi.nlm.nih.govhttp://dx.doi.org/10.1093/jamia/ocz170
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo otaguje tento záznam!

Podobné jednotky