McMurry et al. assess the real-world safety of the BNT162b2 and mRNA-1273 COVID-19 vaccines. Using natural language processing, they compare the rates of specified adverse effects between 68,266 vaccinated individuals and 68,266 matched unvaccinated individuals. They find that both vaccines are safe and tolerated in clinical practice.
actual: 152 words)The recent explosion of biomedical knowledge presents both a major opportunity and challenge for scientists tackling complex problems in healthcare. Here we present an approach for synthesizing biomedical knowledge based on a combination of word-embeddings and select cooccurrences. We evaluated our ability to recapitulate and retrospectively predict disease-gene associations from the Online Mendelian Inheritance in Man (OMIM) resource. Our metrics achieved an area under the curve (AUC) value of 0.981 at the recapitulation task for 2,400 disease-gene associations. At the most stringent cutoff, our metrics predicted 13.89% of these associations before their first cooccurrence in the literature, with a median time of 4 years between prediction and first cooccurrence. Finally, our literature metrics can be combined with human genetics data to retrospectively predict disease-gene associations, IL-6 and Giant Cell Arteritis provided as an example. We believe this framework can provide robust biomedical hypotheses at a much faster pace than current standard practices.
The natural language portions of an electronic health record (EHR) communicate critical information about disease and treatment progression. However, the presence of personally identifying information in this data constrains its broad reuse. In the United States, the Health Insurance Portability and Accountability Act of 1996 (HIPAA) provides a de-identification standard for the removal of protected health information (PHI). Despite continuous improvements in methods for the automated detection of PHI over time, the residual identifiers in clinical notes continue to pose significant challenges - often requiring manual validation and correction that is not scalable to generate the amount of data needed for modern machine learning tools. In this paper, we describe an automated de-identification system that employs an ensemble architecture, incorporating attention-based deep learning models and rule based methods, supported by heuristics for detecting PHI in EHR data. Upon detection of PHI, the system transforms these detected identifiers into plausible, though fictional, surrogates to further obfuscate any leaked identifier. We evaluated the system with a publicly available dataset of 515 notes from the I2B2 2014 de-identification challenge and a dataset of 10,000 notes from the Mayo Clinic. We compared our approach with other existing tools considered best-in-class. The results indicated a recall of 0.992 and 0.994 and a precision of 0.979 and 0.967 on the I2B2 and the Mayo Clinic data, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.