Enhancing Pharmacovigilance with Drug Reviews and Social Media
This work addresses pharmacovigilance for healthcare professionals and researchers by improving ADR detection from unstructured data, but it is incremental as it applies existing BERT methods to new data sources.
This paper tackled the problem of detecting adverse drug reactions (ADRs) by exploring drug reviews and social media as alternative data sources, finding that BERT models, including variants like BioBERT and Clinical BERT, achieved high performance across sentiment classification, ADR presence detection, and named entity recognition tasks.
This paper explores whether the use of drug reviews and social media could be leveraged as potential alternative sources for pharmacovigilance of adverse drug reactions (ADRs). We examined the performance of BERT alongside two variants that are trained on biomedical papers, BioBERT7, and clinical notes, Clinical BERT8. A variety of 8 different BERT models were fine-tuned and compared across three different tasks in order to evaluate their relative performance to one another in the ADR tasks. The tasks include sentiment classification of drug reviews, presence of ADR in twitter postings, and named entity recognition of ADRs in twitter postings. BERT demonstrates its flexibility with high performance across all three different pharmacovigilance related tasks.