Lesion generator for training AI to analyse MRI

Study level

Master of Philosophy


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 human but there is a lack of well characterised large datasets available for training.

This project will focus on image processing techniques to create lesions on healthy scans that could be used for training machine learning methods. This work will fit the activity of a large team working on applying AI to medical imaging.

Research activities

This is a great opportunity for a talented student to work on cutting edge AI techniques as part of a large team of experts.

You will be involved in:

  • organising and managing large magnetic resonance imaging (MRI) datasets
  • developing image processing techniques to generate high resolution realistic lesions on MRI
  • running deep learning technologies to evaluate performance of automated diagnosis.


The outcome of this project will be a new large database that will be used for testing machine learning techniques.

Skills and experience

For this project, we require you to have the following skills:

  • computer science with good programing skills (C++, Python, MATLAB)
  • solid understanding of machine learning
  • interest in deep learning framework such as tensor flow
  • interest and skills in mathematical optimisation and applied mathematics.


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

Annual scholarship round



Contact Professor Olivier Salvado for more information.