Thomas J. Fletcher

Research Associate in Machine Learning at Imperial College London

prof_pic_TF.jpeg

t.fletcher24@imperial.ac.uk

[Google Scholar]

[Github]

Office: I-HUB, Level 5

I am a Research Associate in Machine Learning at Imperial College London, working with Dr Chen Qin in the Department of Electrical and Electronic Engineering and I-X. My current research focuses on motion-robust and efficient fMRI using new techniques such as Gaussian Splatting.

Previously, I was a Research Associate in AI-enabled cardiac MRI at King’s College London in the Cardiac MRI group in the School of Biomedical Engineering and Imaging Sciences. My research there focused on inverse problems and motion correction in MRI reconstruction of highly-undersampled cardiac MRI data.

My undergraduate (Imperial College London), master’s (University College London), and doctoral backgrounds have been in Physics. I was awarded my PhD in Astrophysics from University College London in 2020. For my PhD, I researched the contribution of galaxies towards cosmic reionisation, and the amount of molecular hydrogen in the local Universe using imaging, statistics and large ground- and space-based telescopes. I also worked as a Research Scientist in the Temperature and Humidity Group at the National Physical Laboratory where I worked on thermal imaging projects for ESA and a clinical trial for a thermal imaging device.

news

Sep 27, 2024 I am hiring for a research project on using Gaussian Splatting for Medical Imaging. This will be a short (~9 weeks), part-time role as a Research Assistant at I-X. Please contact me if you are interested.

selected publications

  1. The Lyman continuum escape survey: ionizing radiation from [O III]-strong sources at a redshift of 3.1
    Thomas J Fletcher, Mengtao Tang, Brant E Robertson, Kimihiko Nakajima, Richard S Ellis, Daniel P Stark, and Akio Inoue
    The Astrophysical Journal, 2019
  2. The cosmic abundance of cold gas in the local Universe
    Thomas J Fletcher, Amelie Saintonge, Paula S Soares, and Andrew Pontzen
    arXiv preprint arXiv:2002.04959, 2020
  3. Rapid Motion Estimation and Motion-corrected End-to-end Deep Learning Reconstruction for 1 Heartbeat CINE
    Thomas Fletcher, Gastao Lima Da Cruz, Lina Felsner, Andrew Phair, Haikun Qi, René Botnar, and Claudia Prieto
    Journal of Cardiovascular Magnetic Resonance, 2024