Module overview
This third-year options module introduces students to the principles and analytical methods that underpin precision health, with a focus on modelling clinical risk and uncertainty using statistical and machine learning techniques. Building on foundational biomedical and computational knowledge, students will explore key health datasets, calibrated risk estimation models, and approaches for quantifying and communicating uncertainty in real-world healthcare settings. Through practical examples and state-of-the-art applications, they will learn how bias, missingness, and evidence quality affect model reliability and how uncertainty-aware methods support safe, informed decision-making in precision health.
Aims and Objectives
Learning Outcomes
Knowledge and Understanding
Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:
- Key ML models for personalised risk estimation in health
- Theory of making decision under risk and uncertainty
- Techniques to reduce risks of ML models
- Calibration of ML estimated risks
- The fundamentals of study designs for health innovation and sources of uncertainty
Subject Specific Intellectual and Research Skills
Having successfully completed this module you will be able to:
- Explain the processes relating to risk estimation for a given healthcare problem
- Explain the approaches to modelling person’s decision making in health context
- Critique and compare ML models for risk estimation and calibration
- Apply precision principles for responsible risk estimation so users can make informed decisions
- Identify and utilise key types of health datasets for risk estimation
Transferable and Generic Skills
Having successfully completed this module you will be able to:
- Present and interpret data and risk prediction to/for a non-technical audience
Syllabus
Introduction to the principles of measurement imprecision and uncertainty
Introduction to the principles of the pyramid of evidence for medicine and biology
Introduction to the theory of decision making, including precise and imprecise probabilities
Overview of clinical risks, including incidence, prevalence, absolute vs relative risk
Overview of methods to evaluate the size of the measurement imprecision
ML models for risk estimation, and how they can account for measurement imprecision
ML models calibration, including uncertainty quantification, Platt scaling and isotonic regression
ML models under data complexity, including uncertainty, bias and missingness
Application scenarios and state-of-the-art personalised machine learning models for risk estimation
Learning and Teaching
Teaching and learning methods
The content of this module is delivered through lectures, tutorials and labs, the module website, directed reading and pre-recorded materials.
Students work on their understanding through a combination of independent study and preparation for timetabled activities, along with formative assessments in the form of coursework assignment and exam.
| Type | Hours |
|---|---|
| Wider reading or practice | 25 |
| Preparation for scheduled sessions | 6 |
| Practical classes and workshops | 12 |
| Revision | 17 |
| Completion of assessment task | 42 |
| Lecture | 30 |
| Follow-up work | 18 |
| Total study time | 150 |
Assessment
Summative
This is how we’ll formally assess what you have learned in this module.
| Method | Percentage contribution |
|---|---|
| Coursework | 40% |
| Exam | 60% |
Referral
This is how we’ll assess you if you don’t meet the criteria to pass this module.
| Method | Percentage contribution |
|---|---|
| Coursework | 40% |
| Exam | 60% |
Repeat
An internal repeat is where you take all of your modules again, including any you passed. An external repeat is where you only re-take the modules you failed.
| Method | Percentage contribution |
|---|---|
| Exam | 100% |