Kit Mills Bransby
PhD researcher in Machine Learning and Computer Vision with a focus on Cardiac Imaging applications at Queen Mary University of London.
I am a final year PhD researcher at the school of Electronic Engineering and Computer Science at Queen Mary University of London. Supervised by Dr. Qianni Zhang, Prof. Greg Slabaugh and Prof. Christos Bourantas
My current research focus is the segmentation of cardiac structures using mesh and polygon-based approaches. More broadly I am interested in:
- Graph neural networks
- Combining Point- and dense-based contour representations
- Shortcut learning and out-of-distribution generalizability.
Outside of the lab, I enjoy long-distance running, and all things hilly / green.
news
| Jun 1, 2024 | First author paper accepted at MICCAI 2024: ‘BackMix: Mitigating Shortcut Learning in Echocardiography with Minimal Supervision’. This work took place during an internship at Ultromics. Pre-print available here: PDF |
|---|---|
| May 1, 2024 | I will be part of the delivery team for this years MICCAI workshop of Advances in Simplifying Medical UltraSound (ASMUS). Hope to see you there! |
| Jan 19, 2024 | Starting a 6-month internship as a AI and Computer Vision Researcher at Ultromics, an Oxford-based start up developing AI for echocardiography analysis. |
| May 25, 2023 | First author paper accepted at MICCAI 2023 (early accept!): ‘Joint Dense-Point Representation for Contour-Aware Graph Segmentation’. Paper PDF |
| Jan 19, 2023 | First author paper accepted at ISBI 2023: 3D Coronary Vessel Reconstruction from Bi-Plane Angiography using Graph Convolutional Networks. Paper PDF |
selected publications
-
Joint Dense-Point Representation for Contour-Aware Graph Segmentation(Pre-print) 27th International Conference on Medical Image Computing and Computer Assisted Intervention, 2024 -
Joint Dense-Point Representation for Contour-Aware Graph Segmentation(Pre-print) 26th International Conference on Medical Image Computing and Computer Assisted Intervention, 2023 -
3D Coronary Vessel Reconstruction from Bi-Plane Angiography using Graph Convolutional Networks20th IEEE International Symposium on Biomedical Imaging (ISBI), 2023
