Machine based identification and coding of Adverse Events mentioned in the natural language text received in Pharmacovigilance through the different sources be it emails, faxes, literature, complaint reports, forms, study literatures or phone call transcripts poses a compound problem as it deals with two sub problems, firstly Named entity recognition to detect events (symptoms, illnesses, adverse events) and secondly mapping events to a standard medical coding scheme such as MedDRA. Additionally, industrial applications are required to build their systems in compliance with regulatory requirements, thereby limiting their ability. In this paper we focus on mapping laymen terms or adverse event verbatim/ description to the standard medical terms. We use vector representation of standard medical term dictionary Medical Dictionary for Regulatory Activities (MedDRA) and that of the event verbatim along with their similarity score to establish these mappings.

Medical Coding, Adverse Event Identification, Cosine Similarity, Word2Vec, Semi supervised machine learning

[1] Hazell, Lorna, and Saad AW Shakir. "Under-reporting of adverse drug reactions." Drug safety 29.5 (2006): 385-396.
[2] World Health Organization. "The importance of pharmacovigilance." (2002).
[3] Derczynski, Leon, et al. "Analysis of named entity recognition and linking for tweets." Information Processing & Management51.2 (2015): 32-49.
[4] Lin, Yankai, et al. "Neural relation extraction with selective attention over instances." Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Vol. 1. 2016.
[5] Gupta, Sonal, et al. "Induced lexico-syntactic patterns improve information extraction from online medical forums." Journal of the American Medical Informatics Association 21.5 (2014): 902-909.
[6] Liu, Xiao, and Hsinchun Chen. "AZDrugMiner: an information extraction system for mining patient-reported adverse drug events in online patient forums." International conference on smart health. Springer, Berlin, Heidelberg, 2013.
[7] Wang, Yanshan, et al. "A comparison of word embeddings for the biomedical natural language processing." Journal of biomedical informatics 87 (2018): 12-20.
[8] Pedersen, Ted, et al. "Measures of semantic similarity and relatedness in the biomedical domain." Journal of biomedical informatics 40.3 (2007): 288-299.
[9] Morley, Greg. "Adverse event reporting: A brief overview of MedDRA." Medical Writing 23.2 (2014): 113-116.
[10] T Baldwin, MC de Marneffe, B Han, YB Kim Shared tasks of the 2015 workshop on noisy user-generated text: Twitter lexical normalization and named entity recognition- 2015
[11] Leaman, Robert, et al. "Towards internet-age pharmacovigilance: extracting adverse drug reactions from user posts to health-related social networks." Proceedings of the
2010 workshop on biomedical natural language processing. Association for Computational Linguistics, 2010.
[12] Trask, Andrew, Phil Michalak, and John Liu. "sense2vec-A fast and accurate method for word sense disambiguation in neural word embeddings." arXiv preprint arXiv:1511.06388 (2015).