HAP 464: EHR Configuration & Data Analysis

George Mason University

Retrieval Augmented Generation


In this project you are expected to apply Retrieval Augmented Generation to your project on antidepressants.

Assigned Reading

  • Introduction to retrieval augmented generation (RAG) Read

  • Relevant blogs: RAG optimization Web► Langchain Implementation Web►
  • Relevant Python packages: Langchain Code► llamaindex Code►
  • Generative AI Exists Because of the Transformer Blog►
  • Language Models: A Guide for the Perplexed Read►


Instruction for Submission of Assignments: Assignments should be submitted directly on Blackboard.  In rare situations assignments can be sent directly by email to the instructor. Submission should follow these rules:

  1. Include a terse summary (not more than 1 sentence for each question) at start of your submission or in a separate file.  In the summary indicate if your responses agrees with responses posted to the open web site.
  2. Submit your answers in a Jupyter Notebook Download► YouTube► Slides►
  3. Submit your answers in Blackboard.

Task 1: Using PubMed, identify 10 studies that address effectiveness of antidepressants.  Create summary medication recommendations based on findings of the study using the text within the studies.

Task 2: Use Python to generate text from 500 studies on effectiveness of antidepressants. Include a reference to the original article in these auto-generated summaries.




Copyright 2021 Farrokh Alemi, Ph.D. Most recent revision 02/19/2024.  This page is part of the course on Electronic Health Record Configuration and Data Analysis.