medExtractR: A medication extraction algorithm for electronic health records using the R programming language

Abstract

Objective: We developed medExtractR, a natural language processing system to extract medication dose and timing information from clinical notes. Our system facilitates creation of medication-specific research datasets from electronic health records. Materials and Methods: Written using the R programming language, medExtractR combines lexicon dictionaries and regular expression patterns to identify relevant medication information ('drug entities'). The system is designed to extract particular medications of interest, rather than all possible medications mentioned in a clinical note. MedExtractR was developed on notes from Vanderbilt University's Synthetic Derivative, using two medications (tacrolimus and lamotrigine) prescribed with varying complexity, and with a third drug (allopurinol) used for testing generalizability of results. We evaluated medExtractR and compared it to three existing systems: MedEx, MedXN, and CLAMP. Results: On 50 test notes for each development drug and 110 test notes for the additional drug, medExtractR achieved high overall performance (F-measures > 0.95). This exceeded the performance of the three existing systems across all drugs, with the exception of a couple specific entity-level evaluations including dose amount for lamotrigine and allopurinol. Discussion: MedExtractR successfully extracted medication entities for medications of interest. High performance in entity-level extraction tasks provides a strong foundation for developing robust research datasets for pharmacological research. However, its targeted approach provides a narrower scope compared with existing systems. Conclusion: MedExtractR (available as an R package) achieved high performance values in extracting specific medications from clinical text, leading to higher quality research datasets for drug-related studies than some existing general-purpose medication extraction tools.

A post-processing algorithm for building longitudinal medication dose data from extracted medication information using natural language processing from electronic health records

“…medExtractR MedExtractR is a medication extraction algorithm built using the R programming language, and is more targeted than other more general purpose NLP systems. [8] Given a list of drug names to search for, medExtractR creates a search window around each identified drug mention within a clinical note in which to search for related drug entities.…”

“…[4] Their system extracted medication dosage information with about 80% accuracy. More recently, MedEx [5], CLAMP [6], MedXN [7], and medExtractR [8], have been developed to extract medication information from EHRs more accurately, which is useful for drugbased studies.…”

“…Thus, we developed algorithms that process the raw output from two medication extraction systems, MedXN and medExtractR, which performed best out of 4 NLP systems previously tested. [8] The post-processing algorithm we developed is divided into two parts. Part I processes the raw output from the NLP system and connects entities together.…”

SALTClass: classifying clinical short notes using background knowledge from unlabeled data

“…Among the attempts to develop NLP software for healthcare [3,7,8,9,10,11,12,13,14,15,16], MedLEE (medical language extraction and encoding) [8], cTAKES (clinical text analysis and knowledge extraction system) [9], CogStack [10], and CLAMP [15] are prominent examples. MedLEE is an NLP system that can extract information from textual patient reports based on controlled vocabularies.…”

Text Mining in Healthcare: Bringing Structure to Electronic Health Records

“…Among the attempts to develop NLP software for healthcare (Jonnalagadda et al, 2017;Alex et al, 2019;C. Friedman et al, 2004;Savova et al, 2010;Jackson et al, 2018;Torii et al, 2015;Jonnagaddala et al, 2015;Byrd et al, 2014;Weeks et al, 2019;Soysal et al, 2017;Sohn et al, 2014), medical language extraction and encoding (MedLEE) (C. Friedman et al, 2004), clinical text analysis and knowledge extraction system (cTAKES) (Savova et al, 2010), CogStack (Jackson et al, 2018), and CLAMP (Soysal et al, 2017) are prominent examples. MedLEE is an NLP system that can extract information from textual patient reports based on controlled vocabularies.…”