About

I am an MRes/PhD student in AI-enabled Healthcare at University College London, supervised by Pearse Keane and Daniel Alexander. My project is focusing on the use of machine learning technqiues to study links between retinal images and systemic diseases such as Altzheimer’s disease and strokes.

Recently I have developed an interest in generative methods and their ability to excel in the creation of music based upon motifs and repetition, such as here by Magenta.

Research Experience

My first experience with machine learning models was in quantum chemistry in the Quantum Chemical Topology group at the University of Manchester¹. Here I created simulations of topological atoms and molecules and studied their energy decomposition as they were brought together and compressed.

I then worked on leveraging transfer learning to detect tumours in MRI images of the brain as a research intern in the University of Manchester Computer Science Department². Developing a pretrained variational autoencoder in this project (based on a separate image dataset) prompted me to research further into deep learning, and generative deep learning specifically.

In the research I have participated in so far, I have been fortunate enough to have been exposed to a number of applications of machine learning and I feel that this has illustrated the potential of AI to be applied to a variety of interesting problem areas.

Having learnt about both the rapid progress and present challenges of deep learning and AI more generally, I decided to pursue an MSc at Imperial College London, specializing in AI. At present I am most excited by domain adaptation/transfer learning, especially when applied to data-scarce target domains such as in medical imaging; ideally yielding greater data efficiency and (out of distribution) generalisability.