KIPS Transactions on Software and Data Engineering (정보처리학회논문지:소프트웨어 및 데이터공학)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Detection of Adverse Drug Reactions Using Drug Reviews with BERT+ Algorithm

BERT+ 알고리즘 기반 약물 리뷰를 활용한 약물 이상 반응 탐지

  • 허은영 (한양대학교 컴퓨터소프트웨어학과) ;
  • 정현정 (동덕여자대학교 정보통계학과) ;
  • 김현희 (동덕여자대학교 정보통계학과)
  • Received : 2021.09.30
  • Accepted : 2021.10.26
  • Published : 2021.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we present an approach for detection of adverse drug reactions from drug reviews to compensate limitations of the spontaneous adverse drug reactions reporting system. Considering negative reviews usually contain adverse drug reactions, sentiment analysis on drug reviews was performed and extracted negative reviews. After then, MedDRA dictionary and named entity recognition were applied to the negative reviews to detect adverse drug reactions. For the experiment, drug reviews of Celecoxib, Naproxen, and Ibuprofen from 5 drug review sites, and analyzed. Our results showed that detection of adverse drug reactions is able to compensate to limitation of under-reporting in the spontaneous adverse drugs reactions reporting system.

본 논문에서는 약물의 시판 후 이상 반응을 모니터링하기 위해 약물 리뷰 데이터로부터 약물 이상 반응을 탐지할 수 있는 방법을 제시하였다. 부정적인 약물 리뷰는 주로 약물 이상 반응을 언급하고 있다는 점을 고려하여 약물 리뷰들을 감성 분석하여 부정 리뷰를 추출하고, 부정 리뷰에 사전 기반 추출과 개체명 인식 기법을 적용하여 약물 이상 반응을 탐지하였다. 제안하는 BERT+ 알고리즘으로 부정 리뷰를 판별한 다음, MedDRA 표준 의학 용어 사전을 활용해 이상 반응 단어를 찾고, 개체명 인식 기법을 사용하여 구로 표현된 이상 반응 표현을 탐지하였다. 실험을 위해 비스테로이드성 소염진통제 세 종류의 약물 리뷰를 약물 리뷰 사이트로부터 수집하여 테스트하였으며, 실험 결과는 약물 리뷰를 통한 약물 이상 반응 탐지가 현재의 약물 감시 체계의 한계점을 보완할 수 있음을 보여준다.

Keywords

References

  1. S. M. Rebecca, "Epidemiology of drug allergy," Drug Allergy Test, pp.1-9, 2018.
  2. S. K. Choi, "Understanding of clinical trials and application to the real practice," Korean Journal of Biological Psychiatry, Vol.19, No.4, pp.153-158, 2012.
  3. N. Yoon and M. Kang, "A study of adverse drug event reporting systems in Korea," Korean Journal of Community Pharmacy, Vol.5, No.1, pp.56-65, 2019.
  4. K. Sarvnaz, W. Chen, M-J Alejandro, G. Raj, and P. Cecile,, "Text and data mining techniques in adverse drug reaction detection," ACM Computing Surveys, Vol.47, No.4, pp.1-39, 2015.
  5. X. Wang, G. Hripcsak, M. Markatou, and C. Friedman, "Active computerized pharmacovigilance using matural language processing, statistics, and electronic health records: A feasible study," Journal of American Medical Informatics Association, Vol.16, No.3, pp.328-337, 2009. https://doi.org/10.1016/S1067-5027(09)00041-3
  6. A. Henriksson, M. Kvist, H. Dalianis, and M. Duneld, "Identifying adverse drug event information in clinical notes with distributed semantic representations of context," Journal of Biomeical. Informatics, Vol.57, No.C, pp.333-349, 2015. https://doi.org/10.1016/j.jbi.2015.08.013
  7. A. Sarker, R. Ginn, A. Nikfarjam, K. O'Connor, K. Smith, S. Jayaraman, T. Upadhaya, and G. Gonzalez, "Utilizing social media data for Pharmacovigilance: A review," Journal Biomedical Informatics, Vol.54, pp.202-212, 2015. https://doi.org/10.1016/j.jbi.2015.02.004
  8. Y. Andrew, N. Goharian, and O. Frieder, "Extracting adverse drug reactions from social media," In Proceedings of Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015.
  9. A. Cocos, A. G. Fikes, and A. J. Masino, "Deep learning for pharmacovigilance: Recurrent neural network architectures for labeling adverse drug reactiosn in Twitter posts," Journal of American Medical Informatics Association, Vol.24, No.4, pp.813-821, 2017. https://doi.org/10.1093/jamia/ocw180
  10. A. Nikfarjam, A. Sarker, K. O'connor, R. Ginn, and G. Gonzalez, "Pharmacovigilance from social media: Mining adverse drug reaction mentions using sequence labeling with word embedding cluster features," Journal of American Medical Information Association, Vol.22, No.3, pp.671-681, 2015. https://doi.org/10.1093/jamia/ocu041
  11. F. Graesser, S. Kallumadi, H. Malberg, and S. Zaunseder, "Aspect-based sentiment analysis of drug reviews applying cross-domain and cross-data learning," in Proceedings of the International Conference on Digital Health, pp.121-125, 2018.
  12. I. Korkontzelos, A. Nikfarjam, M. Shardlow, A. Sarker, S. Ananiadou, and G. H. Gonzalez, "Analysis of the effect of sentiment analysis on extracting adverse drug reactions from tweets and forum posts," Journal of Biomedical Informatics, Vol.62, pp.148-158, 2016. https://doi.org/10.1016/j.jbi.2016.06.007
  13. B. Fan, W. Fan, C. Smith, and H. S. Garner, "Adverse drug event detection and extraction from open data: A deep learning approach," Information Processing and Management, Vol.57, No.1, pp.102131, 2020. https://doi.org/10.1016/j.ipm.2019.102131
  14. B. M. R. Spiegel, M. Farid, G. S. Dulai, I. M. Gralnek, and F. Kanwal, "Comparing rates of dyspepsia with Coxibs vs NSAIDS+PPI: A meta-analysis," American Journal of Medicine, Vol.5, No.119, pp.e427-436, 2006.
  15. S. E. Nissen, et al., "Cardiovascular safety of Celecoxib, Naproxen, or Ibuprofen for Arthritis," New England Journal of Medicine, pp.2519-2529, 2016.
  16. MedDRA [Internet], https://www.meddra.org/
  17. A. Cocos, A. G. Fikes, and A. J. Masino, "Deep learning for pharmacovigilance: Recurrent neural network architectures for labeling adverse drug reactions in Twitter posts," Journal of American Medical Information Association, Vol.24, No.4, pp.813-821, 2017. https://doi.org/10.1093/jamia/ocw180
  18. C. S. Wang, P. J. Lin, C. L. Cheng, S. H. Tai, Y. H. K. Yang, B. S. Pharm, and J. H. Chiang, "Detecting potential adverse drug reactions using a deep neural network model," Journal of Medical Internet Research, Vol.21, No.2, pp.e11016, 2019. https://doi.org/10.2196/11016
  19. J. Devlin, M. W. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of deep bidirectional transformers for language understanding," CoRR, https://arxiv.org/abs/1810.04805, 2018.
  20. A. Breden and L. Moore, "Detecting adverse drug reactiosn from twitter through domain-specific preprocessing and BERT ensemble," https://arxiv.org/abs/2005.06634.
  21. S. Hussain, H Afzal, R. Saeed, N. Iltaf, and M. Y. Umaire, "Pharmacovigilance with transformers: A framework to detect adverse drug reactiosn using BERT fine-tuned with FARM," Computational and Mathmatics Methods in Medicine, Vol.2021, https://doi.org/10.1155/2021/5589829
  22. X. Chen, et al., "Mining adverse drug reactions in social media with named entity recognition and semantic methods," Studies in Health Technology and Informatics, Vol.245, pp.322-326, 2017.
  23. J. Lee, W. Yoon, S. Kim, D. Kim, S. Kim, C. H. So, and J. Kang, "BioBERT: A pre-trained biomedical language representation model for biomedical text mining," Bioinformatics, Vol.36, No.4, pp.1234-1240, 2020.
  24. E. Alsentzer, J. Murphy, W. Boag, W-H Weng, D. Jindi, T. Naumann, and M. McDermott, "Publicly available clinical BERT embeddings," In Proceedings of the 2nd Clinical Natural Language Processing Workshop, pp.72-78, 2019.
  25. S. Karimi, A. M-Jimenez, M. Kemp, and C. Wang, "Cadec: A corpus of adverse drug event annotations," Journal of Biomedical Informatics, Vol.55, pp.73-81, 2015. https://doi.org/10.1016/j.jbi.2015.03.010
  26. UCI ML Drug Review Dataset [Internet], https://www.kagle.com/jessicali9530/kuc-hackathon-winter-2018
  27. Z. Huang, W. Xu, and K. Yu, "Bidirectional LSTM-CRF models for sequence tagging," https://arxiv.org/abs/1508.01991/, 2015.
  28. E. Y. Heo, H. Jeong, and H. H. Kim, "A BERT-based labeling technique for drug reviews for signal detection of adverse drug reaction," In Proceedings of Korea Computer Congress, Vol.47, No.2, pp.1394-1396, 2020.
  29. H. Jeong and H. H. Kim, "Detecting and Classification ADRs using Named Entity Recognition on Social Media," In Proceedings of Korea Information Processing Society, Vol.28, No.1, pp.443-446, 2021.
  30. Pennington, Jeffrey, Richard Socher, and Christopher D. Manning. "Glove: Global vectors for word representation," In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp.1532-1543, 2014.