Quantitative Health Sciences Calendar

In medicine, our understanding of diseases, treatments, and preventive measures is perpetually advancing. Despite this, the existing framework of evidence-based medicine (EBM) faces formidable challenges from information overload, as no clinician can catch up with the explosive publication paces, and lacks a robust mechanism for the continuous integration of new evidence. The advent of Generative AI, exemplified by large language models (LLMs) signifies a potentially transformative development in the methodologies employed to collect, appraise, disseminate, and interact with clinical evidence. In this talk, I will discuss our research on addressing the "lost-in-the-middle" issue in evidence retrieval and the fine-tuning of LLMs for effective medical evidence summarization.   

Click here to join or call 1 301 715 8592, Meeting ID: 948 2951 6040 password: 202286

Originally developed in econometrics, SCM creates a weighted combination of control units to estimate what would have happened in the absence of an intervention or exposure. Key conditions for SCM include ensuring no similar interventions are in control units, pre-treatment similarity of covariates, and availability of sufficient pre- and post-treatment data. Real-world applications, including the impact of tobacco taxes on cigarette sales, drug policy changes on overdose rates, and evaluation of HIV prevention programs are discussed.

Click here to join or call 1 301 715 8592, Meeting ID: 948 2951 6040 password: 202286