Skip to main content
. 2022 Sep 14;13:946387. doi: 10.3389/fpsyt.2022.946387

TABLE 9.

Comparison of NLP performance for identification of the best candidates for a specific treatment.

Author Identification NLP model Embeddings and Corpus F1-score [%]
Cohen et al. (101) Candidates for surgery for drug-resistant epilepsy Naïve Bayes ∼6,300 patients’ EHR 82
Sung et al. (103) Intravenous thrombolytic therapy candidates MetaMap-based 234 clinical notes 98.6

NLP, natural language processing; EHR, electronic health record.