I am currently working on motion-robust and efficient fMRI for infants at Imperial within the I-X AI initiative and jointly with Trinity College Dublin. I am using Gaussian splatting and machine learning techniques to reconstruct 4D (3D+time) volumes of the brain from highly motion-corrupted infant fMRI acquisitions.

My role at KCL focused on using machine learning models for inverse problems in cardiac MRI. I developed on a deep learning framework to rapidly reconstruct CINEs (movies of the heart) that are acquired in just 1 heartbeat in the scanner. My work focused on using networks for estimating non-rigid cardiac motion (optical flow) as well as physics-based MRI reconstruction networks. I am developed networks for reconstructing quantitative MRI (cardiac magnetic resonance fingerprinting) using invertible neural networks.

I worked on calculating the amount of ionising radiation escaping from the first galaxies in the Universe. I led an international collaboration of researchers, and using ground and space-based telescopes discovered leaking UV radiation from a large sample of these early galaxies, for the first time. I also used Bayesian statistics and a survey of local galaxies to measure the amount of molecular hydrogen, the fuel for star formation, in the local Universe.

I was part of the Thermal Imaging group, part of the Temperature and Humidity group at NPL. I worked on projects delivering accurate and traceable 3D Thermal Imaging capabilities, including rendering and ray-tracing, for satellites undergoing testing for the European Space Agency's test facility at ESTEC. I also worked as part of a team for a clinical trial comparing a novel thermal imaging device against non-contact infrared thermometers for the treatment of diabetic patients.

I worked on Ion mobility spectrometry (IMS) for a hand-held device for threat detection.

I worked on an internal website to compare benefits of different methods of source verification for clinical trials, using Python and Flask.