Data is Everywhere. Insight is Rare.
From predicting disease outbreaks to identifying patterns in environmental exposures, Machine Learning (ML) is no longer a "future technology"—it is an essential tool for the modern health researcher. This workshop bridges the gap between raw data and actionable public health solutions.
Why this workshop?
- The Power of Prediction: Move beyond traditional statistics to build models that forecast outcomes.
- Competitive Advantage: Gain the technical literacy required for high-impact publishing and modern research grants.
- Transdisciplinary Application: Learn use cases ranging from disease diagnosis to climate-related health risks.
This workshop is taught by three Mount Sinai researchers and clinician-scientists actively publishing in machine learning, causal inference, and natural language processing. You're learning from practitioners, not lecturers.
Vishal Midya, PhD, MSTAT | Lead Facilitator & Instructor
Departments of Environmental Medicine and Public Health
Hachem Saddiki, PhD | Causal Inference in ML
Department of Environmental Medicine and Public Health
Ismail Nabeel, MD, MPH, MS | NLP & Large Language Models in Public Health
Departments of Environmental Medicine and Artificial Intelligence and Human Health
Technical Requirements: You should have some proficiency in R and R Studio.
While we focus on making ML accessible, this is a "keyboard-on" workshop. Participants should have:
Software Proficiency: A working knowledge of R and R Studio.
- Foundational Knowledge: A basic understanding of epidemiology and biostatistics.
- Equipment: A laptop with R andLed by Mount Sinai Faculty Expert
THREE-DAY CURRICULUM
Over three days, you'll progress from foundational statistical ML through specialized applications in public health research.
Day 1: Foundations & Core Methods
Build your technical foundation in regression, classification, and regularization. Learn practical approaches to variable selection, model evaluation, and feature engineering. Work with real-world health datasets in R from the first session.
Topics: Introduction to ML in Public Health · Regression & Classification · Variable Selection & Regularization · Support Vector Machines · Resampling & Feature Engineering · Ensemble Methods
Day 2: Advanced Methods & Domain Applications
Deepen your methodological toolkit with tree-based models, unsupervised learning, and specialized applications. Learn how machine learning intersects with causal inference — a critical skill for health research where understanding why matters as much as prediction.
Topics: Tree-Based Methods & Feature Importance · Unsupervised Learning · Machine Learning in Causal Inference (Theory & Practice) · Introduction to Mixture Models
Day 3: Emerging Methods & Capstone
Explore cutting-edge applications of machine learning in public health, including natural language processing and large language models. Conclude with hands-on risk modeling and a group discussion on future directions in ML for health.
Topics: Natural Language Processing in Public Health · Large Language Models in Health Research · Neural Networks Preliminaries · Risk Scoring & Thresholding · Capstone Discussion: Challenges & Future Directions
LEARNING OUTCOMES
What You'll Be Able to Do
✓ Understand the statistical logic behind modern ML methods — not just which buttons to push, but when and why each technique applies
✓ Build rigorous ML models on health data using industry-standard tools (R, R Studio), from data preparation through evaluation
✓ Navigate the methodological debates in causal inference and predictive modeling that distinguish expert practitioners from cookbook users
✓ Apply specialized techniques like exposure mixture modeling and NLP that most generic ML training skips
✓ Lead independent research using ML as a core method — with confidence in your design choices R Studio pre-installed (setup guide provided upon registration).