TY  - JOUR
ID  - JMK2016
T1  - Cloud–based Evaluation of Organ Segmentation and Landmark Detection Algorithms: VISCERAL Anatomy Benchmarks
A1  - Jimenez del Toro, Oscar
A1  - Taha, Abdel Aziz
A1  - Gruenberg, Katharina
A1  - Krenn, Markus
A1  - Winterstein, Marianne
A1  - Müller, Henning
A1  - Goksel, Orcun
A1  - Menze, Bjoern
A1  - Langs, Georg
A1  - Weber, Marc André
A1  - Fernandez, Tomas Salas
A1  - Foncubierta-Rodríguez, Antonio
A1  - Eggel, Ivan
A1  - Schaer, Roger
A1  - Jakab, András
A1  - Kontokotsios, Georgios
A1  - Dabbah, Mohammad A.
A1  - Dicente Cid, Yashin
A1  - Gass, Tobias
A1  - Heinrich, Mattias
A1  - Jia, Fucang
A1  - Kahl, Fredrik
A1  - Kechichian, Razmig
A1  - Mai, Dominic
A1  - Spanier, Assaf B.
A1  - Vincent, Graham
A1  - Wang, Chunliang
A1  - Hanbury, Allan
JA  - IEEE Transactions on Medical Imaging
Y1  - 2016
UR  - http://ieeexplore.ieee.org/abstract/document/7488206/
M2  - doi: 10.1109/TMI.2016.2578680
N2  - Variations in the shape and appearance of anatomical structures in medical images are often relevant radiological signs of disease. Automatic tools can help automate parts of this manual process. A cloud-based evaluation framework is presented in this paper including results of benchmarking current state-of-the-art medical imaging algorithms for anatomical structure segmentation and landmark detection: the VISCERAL Anatomy benchmarks. The algorithms are implemented in virtual machines in the cloud where participants can only access the training data and can be run privately by the benchmark administrators to objectively compare their performance in an unseen common test set. Overall, 120 computed tomography and magnetic resonance patient volumes were manually annotated to create a standard Gold Corpus containing a total of 1295 structures and 1760 landmarks. Ten participants contributed with automatic algorithms for the organ segmentation task, and three for the landmark localization task. Different algorithms obtained the best scores in the four available imaging modalities and for subsets of anatomical structures. The annotation framework, resulting data set, evaluation setup, results and performance analysis from the three VISCERAL Anatomy benchmarks are presented in this article. Both the VISCERAL data set and Silver Corpus generated with the fusion of the participant algorithms on a larger set of non-manually-annotated medical images are available to the research community.
ER  -