Data augmentation for training deep learning networks to detect brain lesions from MRI

Study level


Topic status

We're looking for students to study this topic.


Professor Clinton Fookes
Division / Faculty
Science and Engineering Faculty


Detecting lesions from medical imaging is a difficult tasks for radiologists that is time-consuming and subjective. Artificial intelligence (AI) techniques could outperform humans but there is a lack of well-characterised, large datasets available for training purposes.

This PhD project will focus on data augmentation techniques using synthetic approaches, as well as weekly supervised learning.

This project is part of a large team of researchers involving startups, CSIRO, QUT faculties and several postdoc and PhD students.

Research activities

This project is an opportunity for you to work on innovative AI techniques as part of a large team of experts while being involved in a start-up environment in close collaboration with clinicians.

You will be involved in organising and managing large MRI datasets and developing image processing techniques to generate high resolution realistic lesions on a 3D brain MRI. These MRI scans will span a range of ages, shapes and MR contrasts.

Techniques and experiments will be based around data augmentation and deep learning-based synthesis approaches and may include physics-based constraints.


The results of the project will be communicated with the team and collaborators and presented at international conferences and in peer-reviewed journals.


You may be able to apply for a research scholarship in our annual scholarship round.

Annual scholarship round



Contact Olivier Salvado for more information