HDR UK Site
The aim of the project at King’s College London was to evaluate machine learning algorithms developed for cardiac magnetic resonance imaging (CMR), trained on UK Biobank, on patient data in clinical CMR centres linked with HDR sites (e.g. St Thomas’, Barts BioResource).
In the short duration of the postdoc working on this grant, Dr Ines Machado (6 months until 12/2020), advances have been made for data from both UK biobank, with preliminary testing on patient data acquired at St. Thomas’ Hospital trust, resulting in a publication to the prestigious MICCAI conference satellite event STACOM. She also contributed to a second paper for that event, on cardiac MR image domain generalisation. Journal paper submissions for both double-blind, peer-reviewed, full-length conference publications are in preparation.
Project Team and Collaborators
- Professor Julia Schnabel, King’s College London and Helmholtz Center Munich & Technical University of Munich – project lead
- Dr Ines Machado, King’s College London – funded by HDR UK
- Dr Andrew King, King’s College London
- Prof Alistair Young, King’s College London
- Prof Claudia Prieto, King’s College London
- Prof Miguel Castelo-Branco, University of Coimbra
- Dr Devran Ugurl, King’s College London
- Dr Gastao Cruz, King’s College London
- Dr Bram Ruijsink, King’s College London and St. Thomas’ Hospital Trust
- Dr Kerstin Hammernik, Imperial College London and Technical University of Munich
- Steffen Petersen, NIHR Barts Biomedical Research Centre (BRC)
deep learning for image reconstruction; accelerated MRI; image segmentation; quality assessment; domain generalisation
CogStack information retrieval and extraction platform gives access to underused data
27 February 2023
CogStack represents a major advance in the capacity to extract and analyse unstructured data from electronic health records (EHRs). It uses a range of technologies to support modern open-source...
Biobank conversion allows greater UK contribution to international research
21 February 2023
A team of researchers processed and converted over 1.3 billion rows of UK Biobank data to the Observational Medical Outcomes Partnership (OMOP) common data model (CDM), improving its usability for...