A5018938525- Health disparities and outcomes
- Advanced Causal Inference Techniques
- Topic Modeling
- Vaccine Coverage and Hesitancy
- Machine Learning in Healthcare
- COVID-19 epidemiological studies
Boston University
2020
Purpose: Extracting and structuring relevant clinical information from electronic health records (EHRs) remains a challenge due to the heterogeneity of systems, documents, documentation practices. Large Language Models (LLMs) provide an approach processing semi-structured unstructured EHR data, enabling classification, extraction, standardization. Methods: Medical documents are processed through structured data pipeline generate normalized FHIR data. Unstructured undergoes preprocessing,...
Systems models, which by design aim to capture multi-level complexity, are a natural choice of tool for bridging the divide between social epidemiology and causal inference. In this commentary, we discuss potential uses complex systems models improving our understanding quantitative effects in epidemiology. To put context, will describe how approach could be used optimise distribution COVID-19 response resources minimise inequalities during after pandemic.
Background: Developing a causal graph is an important step in etiologic research planning and can be used to highlight data flaws irreparable bias confounding. As case study, we consider recent findings that suggest human papillomavirus (HPV) vaccine less effective against HPV-associated disease among girls living with HIV compared without HIV. Objectives: To understand the relationship between status HPV effectiveness, it outline key assumptions of mechanisms before designing study...