PH.140.850 - How to AI (for public health)

A course on applying AI tools and methods to public health challenges

First Term 2025
Johns Hopkins Bloomberg School of Public Health

Instructor: Yiqun T. Chen
Office: Wolfe Street Building, E3608
Office Hours: Schedule via email
E-mail: yiqunc@jhu.edu

Course Meeting Times: Mondays 3:30–4:50 pm
Location: Wolfe Street Building, W2015

Anonymous Feedback: link

Course Description

Recent advances in Artificial Intelligence (AI), such as ChatGPT, have shown impressive capabilities in language understanding and generation, demonstrating both fluency and broad knowledge retention. As these tools promise to transform evidence synthesis, idea generation, and the design of innovative solutions, AI is emerging as a pivotal technology for researchers and practitioners across disciplines.

This course introduces students to the practical applications of AI in public health and medicine. Students will use data-driven methods to understand, predict, and address public health challenges. Topics include a high-level overview of modern AI models, with emphasis on their integration into health domains, as well as critical engagement with the ethical, legal, and societal issues related to their deployment. Each class will combine lecture and discussion, including brainstorming sessions on how AI could power and advance students’ research and practice.

Intended Audience: Master’s and PhD students in Biostatistics or other departments with quantitative training, or by instructor permission.

Prerequisites: Familiarity with basic statistics and an interest in biomedical/public health applications are expected. Experience with Python (or R) and coursework in statistical modeling or machine learning are recommended, but not required.

Resources

Course Materials: Readings, lecture slides, and coding notebooks will be posted on this website.

Schedule

Week Date Topic Materials
1 Aug 25 Brief Review of Modern AI and ML Slides (PDF)
- Sep 1 Holiday (Labor Day) - No class -
2 Sep 8 Foundations: Data structure, feature representation, and practical AI tools Slides (PDF)
- Sep 15 Class cancelled - Makeup on Oct 3 (Friday) -
3 Sep 22 Continue on word embeddings and intro to large language models Slides (PDF)
4 Sep 29 Large language models Annotated Slides; Demo Code
5 Oct 3 (Fri) Makeup class: More LLMs and Intro to Multimodal Models Slides (PDF)
6 Oct 6 Vision-Language Models, Part II Annotated Slides
7 Oct 13 Agents and Tool-Augmented Workflows Tentative Slides
8 Oct 20 How is AI changing Public Health Data Science TBA

Project

Evaluation and Grading: Pass/Fail. Choose one of the two tracks below, complete the deliverable, and email it to yiqunc@jhu.edu by Friday, Oct 24 at 11:59 pm ET. Grades will be entered by Oct 27; reach out before the deadline if you need more time.

Track A: Applied AI Project

  • Analyze any real-world dataset of interest using modern AI/ML methods.
  • Document your workflow with exploratory analysis, model iterations, and key takeaways in a write-up.
  • The write-up should be at least two pages long with three sections at minimum: 1. introduction of the problem; 2. how you framed it as an ML/AI problem; and 3. results on applying the methods on your data. You should include at least one figure or table for your results, and articulate what metircs, predictors, and outcomes you have.

Track B: Mathematical Foundations

  • Work through the theory problem set and submit solutions to at least five with good-faith attempts. You only need five to pass.
  • Write clearly, show intermediate steps, and state any assumptions or references you use.
  • Resources: Theory Questions (PDF) · Theory Questions (TeX)

Submission Checklist

  • Label your file with your name and track (e.g., Firstnamr_Lastname_TrackA.pdf or Firstnamr_Lastname_TrackB.pdf).
  • Email the deliverable to yiqunc@jhu.edu by 11:59 pm ET on Oct 24.
  • Optional: share drafts or questions during office hours for feedback before the deadline.

Learning Objectives

Upon completion of this course, students will be able to:

  • Identify major AI methods used in public health
  • Apply AI tools to real-world data science problems
  • Evaluate ethical, social, and fairness considerations in AI applications to population health