We live in a world where health data comes in many complex forms where traditional statistical approaches may be unable to be applied, for example image, text or signal data. Complex machine learning models show great promise in health sciences, for example in detection of cancers in MRI imaging. How do these machine learning models work, and how are they different to traditional statistical modelling?
This module will provide an understanding of machine learning approaches to data science and teach the necessary skills to build and evaluate models on real world data. Students will learn traditional “shallow” learning algorithms, as well as state of the art deep learning approaches; and be able to make an informed decision on which is appropriate for the problem at hand. With a focus on explainability, interpretability and clinical usability, we will focus not only on how to build these models, but how they can be used clinically to maximise impact in a healthcare environment.
Students will benefit from working alongside leading researchers in the field and develop hands-on experience building data models using real health datasets; reviewing state of the art research and implementing new methodologies using relevant data.