Petru-Daniel Tudosiu
EPSRC funded PhD student working on Hetero-multi-modal stroke diagnosis and prognosis deep neural network model. My research focuses on developing full 3D deep learning generative models of neurological pathologies that are reliable and reproducible. The practical aim is to use their disentangled latent representation for prognosis and diagnosis of stroke patients. The personal aim is to release well documented and stable networks that people will enjoy to use and expand upon. Before starting my PhD I have received my B.Sc degree in Computer Science from University of Southampton with a focus classical Machine Learning. Following that, I studied Computational Biology at University College London investigating the mechanistic behaviour of Delirium in a data-driven way.
Projects
Radiology - TriagingGenerative Modelling
Publications
Neuromorphologicaly-preserving Volumetric data encoding using VQ-VAE
Tudosiu, P.D., Varsavsky, T., Shaw, R., Graham, M., Nachev, P., Ourselin, S., Sudre, C.H. and Cardoso, M.J., 2020. arXiv preprint arXiv:2002.05692.