“Can we build a machine capable of interpreting medical scans
with super-human performance?”
I am Senior-Lecturer in Medical Image Computing at Imperial College London and a member of the Biomedical Image Analysis Group. I am also the Champion for Public Engagement in the Department of Computing.
1st Place at 2017 BraTS Challenge
Our BioMedIA team has won the 2017 Brain Tumour Segmentation Challenge (BraTS). We have developed a deep learning approach, called EMMA, which stands for Ensembles of Multiple Models and Architectures, which obtained best performance among more than 50 participating teams
DeepMedic Source Code Released
Just released our DeepMedic software on Github. That’s deep, fully 3D convolutional neural networks for medical image segmentation! The code has been used in our winning entry for the ISLES 2015 stroke lesion segmentation challenge, and for brain tumour segmentation and traumatic brain injuries.
NVIDIA Global Impact Award Finalists
We are very excited to be placed among five finalists for the 2016 NVIDIA Global Impact Award with our work on Deep Learning for Brain Lesion Segmentation. The credit goes to PhD student Konstantinos Kamnitsas. More details about the work can be found here.
2015 Imperial Festival Demo
ElasticFusion presented at RSS 2015
The BioMedIA group is a member of the EPSRC-NIHR Healthcare Technology Co-operatives Partnership Medical Image Analysis Network (MedIAN) which seeks to bring together leading UK researchers in medical image analysis with relevant stakeholders to identify new opportunities for medical image analysis methodology research and early clinical translation. More information can be found on the MedIAN website.
As a founding member of an international research initiative on spine imaging, we are proud to announce the official opening of the SpineWeb – A Collaborative Platform for Research on Spine Imaging and Image Analysis. Researchers are invited to actively participate in this initiative and contribute to the SpineWeb platform.