Abstract | ||
---|---|---|
•Phenotyping of opioid overdose cases stratified by severity using machine learning.•Random forests were superior to all other methods (AUC = 0.893).•Features derived from the OMOP CDM and NLP boost performance.•Ordinal models were inferior to traditional classification methods. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.jbi.2019.103185 | Journal of Biomedical Informatics |
Keywords | Field | DocType |
Machine learning,Opioid,Phenotype,Overdose,Electronic health record | Population,Receiver operating characteristic,Computer science,Opioid overdose,Disparate system,Chart,Artificial intelligence,Random forest,Logistic regression,Medical diagnosis,Machine learning | Journal |
Volume | ISSN | Citations |
94 | 1532-0464 | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jonathan Badger | 1 | 0 | 0.34 |
Eric LaRose | 2 | 10 | 1.61 |
John Mayer | 3 | 6 | 1.63 |
Fereshteh Bashiri | 4 | 0 | 0.34 |
David Page | 5 | 533 | 61.12 |
Peggy Peissig | 6 | 189 | 23.83 |