OVERVIEW
Research with the ITMAT Data Science Group
Current research within the ITMAT Data Science Group focuses on new cross-domain challenging projects such as deep-learning with medical images (diagnosis, lesion quantification, predictive modeling), multi-omics data association, disease subtyping, unsupervised learning on electronic health-records.
The focus of the ITMAT Data Science Group is to enable biomedical scientific discovery through advanced applications of computational methods, development of novel algorithms and software, curation and sharing of analytical pipelines and data, and maintenance of existing software tools. The group has expertise in state of the art data analytics and in the use of HPC at Imperial.
Productivity and Sustainability in clinical data analysis to enable rapid translation is the key goal of the group.
Research with international collaborators
Very active collaborators with Columbia University (USA), Telecom ParisTech, Institut Pasteur, Arts et Metiers ParisTech (France) and with the industry (GE, Siemens, Echosens,...).
Guest Lectures
Keynote: "Unsupervised Learning and Domain Adaptation for Subtyping of Emphysema on Very Large Cohorts of Lung CT Scans", International Conference on Medical Imaging with Deep Learning (MIDL), Online, 2020