In this module, students with basic knowledge of machine learning (ML) will learn in-depth ML algorithms and data analysis focusing on recent techniques, such as deep learning with neural networks (convolutional and recurrent neural networks). Methods in supervised learning, unsupervised learning and reinforcement learning are introduced in the course as well as practical techniques and tips to effectively build and maintain codes for various applications. This is an intensive, practical course for people wanting to gain hands-on experience of modelling and analytic ML techniques for healthcare.
Face to Face Dates
Apr: 13, 20, 27
May:04, 11, 18, 25
Jun: 01, 08, 15
This module outlines the definition of machine learning and its application in healthcare and biomedicine. Foundational concepts of probabilistic learning, decision tree learning, artificial neural networks, support vector machines, reinforcement learning, ensemble classifier, and deep learning are applied to datasets in healthcare. Key concepts of dimensionality reduction are applied to health data. Formative assessment is conducted during the learning process to improve students’ attainment. A guided discussion of latest publications within small groups supports the idea of peer-assessment; it is an educational practice in which students interact with each other to attain and assess educational goals. Moreover, mini-projects under the guidance of the course facilitator help to assess the ability to transfer theoretical knowledge into practical implementations.
Teaching and learning methods
The module comprises ten sessions of learning. By the end of the module, students should understand and be able to communicate the benefits of:
Building models using ML algorithms, writing practical codes, optimising codes;
At the end of the module students should be able to:
Be familiar with widely-used ML algorithms;
Create structured code for practical implementation and reproducible research;
Know the general approaches to optimize the models;
Understand how to build advanced ML models to solve supervised, and unsupervised learning problem;
Understand how does reinforcement learning work, and how to implement it in practice;
Understand the mathematics necessary for constructing ML solutions;
Work with a range of dataset, e.g. labelled data, clinical data, time series data, etc;
Be able to design and implement various ML algorithms in a range of real-world applications;
The summative assessment refers to the outcome of the entire module and culminates in a documented paper of 2000-2500 words and the submission of an iPython-notebook.