Résumé en anglais

Kidney segmentation in 3D CT images allows extracting useful information for nephrologists. For practical use in clinical routine, such an algorithm should be fast, automatic and robust to contrast-agent enhancement and elds of view. By combining and re ning state-of-the-art techniques (random forests and template deformation), we demonstrate the possibility of building an algorithm that meets these requirements. Kidneys are localized with random forests following a coarse to fi ne strategy. Their initial positions detected with global contextual information are re ned with a cascade of local regression forests. A classi cation forest is then used to obtain a probabilistic segmentation of both kidneys. The nal segmentation is performed with an implicit template deformation algorithm driven by these kidney probability maps. Our method has been validated on a highly heterogeneous database of 233 CT scans from 89 patients. 80 % of the kidneys were accurately detected and segmented (Dice coe cient > 0:90) in a few seconds per volume.