T
Tom Vercauteren
Researcher at King's College London
Publications - 443
Citations - 19699
Tom Vercauteren is an academic researcher from King's College London. The author has contributed to research in topics: Segmentation & Computer science. The author has an hindex of 47, co-authored 381 publications receiving 14216 citations. Previous affiliations of Tom Vercauteren include Mauna Kea Technologies & Wellcome Trust.
Papers
More filters
Learning joint lesion and tissue segmentation from task-specific hetero-modal datasets
TL;DR: This work proposes a novel approach to build a joint tissue and lesion segmentation model from task-specific hetero-modal and partially annotated datasets and exploits an upper-bound of the risk to deal with missing imaging modalities.
Book ChapterDOI
Incompressible Image Registration Using Divergence-Conforming B-Splines
Lucas Fidon,Michael Ebner,Luis C. Garcia-Peraza-Herrera,Marc Modat,Sebastien Ourselin,Tom Vercauteren +5 more
TL;DR: Accuracy measurements demonstrate that the proposed SVFs framework compares favourably with state-of-the-art methods whilst achieving volume preservation, and the numerical incompressibility error for the transformation in the case of an Euler integration is studied.
Book ChapterDOI
Conditional Segmentation in Lieu of Image Registration
TL;DR: In this paper, the location of corresponding image-specific ROIs, defined in one image, within another image is learned by a conditional segmentation algorithm, which can build on typical image segmentation networks and their widely adopted training strategies.
Journal ArticleDOI
Randomized phase 2 trial of adaptive dose painting vs. standard IMRT for head and neck cancer
Fréderic Duprez,J. Daisne,Dieter Berwouts,Werner De Gersem,Ingeborg Goethals,L. Olteanu,Tom Vercauteren,Wilfried De Neve +7 more
Journal ArticleDOI
Zero-shot super-resolution with a physically-motivated downsampling kernel for endomicroscopy
Agnieszka Barbara Szczotka,Dzhoshkun I. Shakir,Matthew J. Clarkson,Stephen P. Pereira,Tom Vercauteren +4 more
TL;DR: In this article, a zero-shot super-resolution (ZSSR) approach was proposed to improve the quality of endomicroscopy images without the need for ground truth HR images.