UBC DASH PRESENTS MEDI 504B
January 5, 2026 - February 11, 2026 | M/W; 1pm - 3pm, In-person at DHCC 10207
MEDI504 is a Special Topics course designed to develop literacy and conceptual understanding of data science to enable effective collaboration with data scientists as related to health research spanning the basic, clinical, population and health services.
Course Information
Dates: January 5, 2026 - February 11, 2026
Time: Monday and Wednesday; 1:00pm - 3:00pm
Location: In-person at DHCC 10207 (2775 Laurel St, Vancouver, BC V5Z 1M9, Room 10207)
Credit: 1.5
Pre-requisite: STAT 545A Exploratory Data Analysis
*Pre-requisite exemption requests will be reviewed on a case-by-case basis. For more information, please contact the program office (medi.exptlmed@ubc.ca).
Learning Outcomes
This course targets students who are interested in how medical care can be transformed by “Big data”. The instructors will use real-life clinical use cases to motivate how clinical care can be enhanced by collecting unconventional data of various kinds and feeding such data to machine learning methods to build next-generation clinical tools. However, even with the appropriate data collected, the teaching team will describe the data science challenges that need to be overcome before the data can be used properly. This course will introduce the main concepts and methods in biostatistics, and machine learning, including unsupervised and supervised methods, and data visualization. Last but not least, there will be discussion of ethical and fairness issues, and other potential pitfalls of applying machine learning to medical care.
The four major learning outcomes for the students are:
- Understand the fundamental concepts of machine learning and predictive modeling.
- Apply data science processes and project management techniques to biomedical datasets.
- Perform exploratory data analysis (EDA) and create user stories for data science projects.
- Develop and evaluate logistic regression models and other machine learning techniques.
- Address challenges such as missing data, class imbalance, and model overfitting.
- Gain experience with advanced techniques including regularization, boosting, and deep learning.
- Critically assess model discrimination, calibration, and net benefit.
Topics Schedule
Week 1 | Foundations of Predictive Modelling EDA in Practice and User Stories |
Week 2 | Logistic Regression Missing Data & Class Imbalance |
Week 3 | Feature Selection Regularization |
Week 4 | Trees & Forests Boosting |
Week 5 | Discrimination & Calibration Net Benefit |
Week 6 | Deep Learning Convolutional Neural Networks |
Course Instructors
Dr. Aline Talhouk, Dr. Min Hyung Ryu
Assistant Professor Assistant Professor
Obstetrics & Gynecology Respiratory Medicine