Machine learning-based identification and rule-based normalization of adverse drug reactions in drug labels

Abstract

BackgroundUse of medication can cause adverse drug reactions (ADRs), unwanted or unexpected events, which are a major safety concern. Drug labels, or prescribing information or package inserts, describe ADRs. Therefore, systematically identifying ADR information from drug labels is critical in multiple aspects; however, this task is challenging due to the nature of the natural language of drug labels.ResultsIn this paper, we present a machine learning- and rule-based system for the identification of ADR entity mentions in the text of drug labels and their normalization through the Medical Dictionary for Regulatory Activities (MedDRA) dictionary. The machine learning approach is based on a recently proposed deep learning architecture, which integrates bi-directional Long Short-Term Memory (Bi-LSTM), Convolutional Neural Network (CNN), and Conditional Random Fields (CRF) for entity recognition. The rule-based approach, used for normalizing the identified ADR mentions to MedDRA terms, is based on an extension of our in-house text-mining system, SciMiner. We evaluated our system on the Text Analysis Conference (TAC) Adverse Drug Reaction 2017 challenge test data set, consisting of 200 manually curated US FDA drug labels. Our ML-based system achieved 77.0% F1 score on the task of ADR mention recognition and 82.6% micro-averaged F1 score on the task of ADR normalization, while rule-based system achieved 67.4 and 77.6% F1 scores, respectively.ConclusionOur study demonstrates that a system composed of a deep learning architecture for entity recognition and a rule-based model for entity normalization is a promising approach for ADR extraction from drug labels.

ADR databases for on‐site clinical use: Potentials of summary of products characteristics

“…In a paper by Ly, three different NLP systems were used for the extraction of ADR terms from SPL for 10 products with F1 scores from 0.67 to 0.79 23 . Tiftikci used machine learning and rule‐based systems for ADR identification in SPLs for 99 products with a F1 score of 0.77 24 . Pandey described the use of one NLP system to extract ADR terms from SPL with a F1 score of 0.79 for a total of 100 SPLs 25 .…”

Aspects of Utilization and Limitations of Artificial Intelligence in Drug Safety

“…discharge summaries). We could automate medical dictionary for regulatory activities (MedDRA) coding and the WHO drug dictionary coding using ML along with rule-based systems [10]. We could use AI along with rule-based systems to check for duplicates and for categorizing ICSRs (e.g.…”

A 2018 workshop: vaccine and drug ontology studies (VDOS 2018)

“…This year, four full-length papers and one short-length paper were accepted for oral presentations at the workshop after a peer-review process with each submission reviewed by at least three independent reviewers. After one additional round of independent peer reviewing on their extended version, with the reviewers' comments taken care of, by the workshop co-organizers and the journal editors, four full-length papers [40][41][42][43] have been accepted for publication in the current thematic issue of the BMC Bioinformatics.…”

“…Lastly, Tiftikci et al [43] presented a machine learning (ML)-and rule-based system for identifying adverse drug reaction (ADR) mentions in the text of drug labels and their normalization through the Medical Dictionary for Regulatory Activities (MedDRA) dictionary. ADRs, unwanted or unexpected events from using drugs, are a major safety concern, and drug labels describe established ADRs for the given drug.…”