21 March 2016 A learning-based, fully automatic liver tumor segmentation pipeline based on sparsely annotated training data
Author Affiliations +
Abstract
Current fully automatic liver tumor segmentation systems are designed to work on a single CT-image. This hinders these systems from the detection of more complex types of liver tumor. We therefore present a new algorithm for liver tumor segmentation that allows incorporating different CT scans and requires no manual interaction. We derive a liver segmentation with state-of-the-art shape models which are robust to initialization. The tumor segmentation is then achieved by classifying all voxels into healthy or tumorous tissue using Extremely Randomized Trees with an auto-context learning scheme. Using DALSA enables us to learn from only sparse annotations and allows a fast set-up for new image settings. We validate the quality of our algorithm with exemplary segmentation results.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michael Goetz, Michael Goetz, Eric Heim, Eric Heim, Keno Maerz, Keno Maerz, Tobias Norajitra, Tobias Norajitra, Mohammadreza Hafezi, Mohammadreza Hafezi, Nassim Fard, Nassim Fard, Arianeb Mehrabi, Arianeb Mehrabi, Max Knoll, Max Knoll, Christian Weber, Christian Weber, Lena Maier-Hein, Lena Maier-Hein, Klaus H. Maier-Hein, Klaus H. Maier-Hein, } "A learning-based, fully automatic liver tumor segmentation pipeline based on sparsely annotated training data", Proc. SPIE 9784, Medical Imaging 2016: Image Processing, 97841I (21 March 2016); doi: 10.1117/12.2217655; https://doi.org/10.1117/12.2217655
PROCEEDINGS
6 PAGES


SHARE
Back to Top