Header

UZH-Logo

Maintenance Infos

Categorization of free-text drug orders using character-level recurrent neural networks


Raiskin, Yarden; Eickhoff, Carsten; Beeler, Patrick E (2019). Categorization of free-text drug orders using character-level recurrent neural networks. International Journal of Medical Informatics, 129:20-28.

Abstract

Background and purpose Manual annotation and categorization of non-standardized text (“free-text”) of drug orders entered into electronic health records is a labor-intensive task. However, standardization is required for drug order analyses and has implications for clinical decision support. Machine learning could help to speed up manual labelling efforts. The objective of this study was to analyze the performance of deep machine learning methods to annotate non-standardized text of drug order entries with their therapeutically active ingredients.
Materials and methods The data consisted of drug orders entered 8/2009-4/2014 into the electronic health records of inpatients at a large tertiary care academic medical center. We manually annotated the most frequent order entry patterns with the active ingredient they contain (e.g. “Prograf”⟵“Tacrolimus”). We heuristically included additional orders by means of character sequence comparisons to augment the training dataset. Finally, we trained and employed character-level recurrent deep neural networks to classify non-standardized text of drug order entries according to their active ingredients.
Results A total of 26,611 distinct order patterns were considered in our study, of which the top 7.6% (2028) had been annotated with one of 558 distinct ingredients, leaving 24,583 unlabeled observations. Character-level recurrent deep neural networks achieved a Mean Reciprocal Rank (MRR) of 98% and outperformed the best representative baseline, a trigram-based Support Vector Machine, by 2 percentage points.
Conclusion Character-level recurrent deep neural networks can be used to map the active ingredient to non-standardized text of drug order entries, outperforming other representative techniques. While machine learning might help to facilitate categorization tasks, still a considerable amount of manual labelling and reviewing work is required to train such systems.

Abstract

Background and purpose Manual annotation and categorization of non-standardized text (“free-text”) of drug orders entered into electronic health records is a labor-intensive task. However, standardization is required for drug order analyses and has implications for clinical decision support. Machine learning could help to speed up manual labelling efforts. The objective of this study was to analyze the performance of deep machine learning methods to annotate non-standardized text of drug order entries with their therapeutically active ingredients.
Materials and methods The data consisted of drug orders entered 8/2009-4/2014 into the electronic health records of inpatients at a large tertiary care academic medical center. We manually annotated the most frequent order entry patterns with the active ingredient they contain (e.g. “Prograf”⟵“Tacrolimus”). We heuristically included additional orders by means of character sequence comparisons to augment the training dataset. Finally, we trained and employed character-level recurrent deep neural networks to classify non-standardized text of drug order entries according to their active ingredients.
Results A total of 26,611 distinct order patterns were considered in our study, of which the top 7.6% (2028) had been annotated with one of 558 distinct ingredients, leaving 24,583 unlabeled observations. Character-level recurrent deep neural networks achieved a Mean Reciprocal Rank (MRR) of 98% and outperformed the best representative baseline, a trigram-based Support Vector Machine, by 2 percentage points.
Conclusion Character-level recurrent deep neural networks can be used to map the active ingredient to non-standardized text of drug order entries, outperforming other representative techniques. While machine learning might help to facilitate categorization tasks, still a considerable amount of manual labelling and reviewing work is required to train such systems.

Statistics

Citations

Altmetrics

Downloads

3 downloads since deposited on 19 Sep 2019
3 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > University Hospital Zurich > Clinic and Policlinic for Internal Medicine
07 Faculty of Science > Institute of Mathematics
Dewey Decimal Classification:610 Medicine & health
Uncontrolled Keywords:Health Informatics
Language:English
Date:1 September 2019
Deposited On:19 Sep 2019 12:55
Last Modified:25 Sep 2019 00:43
Publisher:Elsevier
ISSN:1386-5056
OA Status:Closed
Publisher DOI:https://doi.org/10.1016/j.ijmedinf.2019.05.020
PubMed ID:31445256
Project Information:
  • : FunderSNSF
  • : Grant IDPZ00P2_174025
  • : Project TitleRepresentation Learning for Clinical Artificial Intelligence

Download

Closed Access: Download allowed only for UZH members

Content: Accepted Version
Filetype: PDF - Registered users only until 23 May 2020
Size: 308kB
Embargo till: 2020-05-23