Machine Learning in Healthcare

BST 209 Subject & Catalog Number

Description

Machine learning (ML) is revolutionizing healthcare, from predictive analytics and medical imaging to personalized medicine and AI-driven diagnostics. This course provides students with an applied understanding of machine learning (ML) and deep learning (DL) techniques in healthcare, focusing on practical coding in R and the latest innovations in large language models (LLMs) and AI regulatory landscapes.

Through a mix of lectures, hands-on coding exercises, and discussions on cutting-edge AI advancements, students will learn to build and evaluate ML models for real-world biomedical applications. Topics include ML terminology and workflow, ML performance metrics, supervised ML (regression, tree-based models), unsupervised ML (clustering, dimensionality reduction), deep learning (CNNs for medical imaging), model selection, and model optimization.

A hands-on group project will be introduced in the first week, and students will work on it throughout the course, progressively applying new concepts and refining their models. Teams will analyze a real-world healthcare dataset, build ML models, and interpret their findings. Regular check-in sessions will provide feedback and guidance, culminating in a final presentation where students showcase their results, model evaluations, and insights.

The course will also include exposure to the evolving landscape of ML/DL/LLMs in healthcare, including AI regulation, clinical validation, and the integration of AI into healthcare systems.
 
Students are expected to have R and RStudio installed on their laptops and to be familiar with programming in R. The instructors highly recommend completing an online introduction to R course (via edX, Coursera, Data Camp, Udacity, etc.) before enrolling in this course if the student hasn’t been exposed to R. A brief review of programming in R will take place during the first session.