RV Segmentation Challenge in Cardiac MRI
Evaluation of right ventricular structure and function is of importance in the management of most cardiac disorders . RV imaging has always been considered challenging, mainly because of the complex motion and anatomy of the RV. Indeed RV segmentation in MR images involves many problems such as dealing with the highly variable, crescent shape of the RV and its thin and ill-defined borders. In the literature, several segmentation methods have been proposed, relying in particular on statistical shape modeling and/or 3D approaches, in order to deal with the aforementioned issues . Because many segmentation algorithms in cardiac MRI jointly segment both ventricles, the database contains images with both ventricles clearly visible.
Whereas the left ventricle (LV) segmentation in cardiac MRI has given rise to two segmentation challenges (part of MICCAI'09 and MICCAI'11), RV segmentation algorithms have never been assessed on common data. The aim of this challenge is to gather researchers in this field to advance towards clinical applications.
Our challenge was part of a workshop 3D Cardiovascular Imaging: a MICCAI segmentation challenge at the 15th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), which was held on October 1st, 2012 in Nice Sophia Antipolis, France. This workshop also includes a challenge on Coronary Artery Stenoses Detection and Quantification.
Outline of the MICCAI'12 Right Ventricle Segmentation Challenge
(Oct 1st, 2012)
The goal of this contest is to compare right ventricle (RV) segmentation methods by providing a common database of cardiac cine MR images and associated expert contours, as well as an evaluation system. Challengers are invited to enter their segmentation algorithm to automatically find RV endocardium and epicardium, with little or no user intervention.
Participants are provided with a Training Set (images+ground truth) and a Test1 Set (images only). They are invited to submit a paper describing their methods and results to our MICCAI'12 challenge (see How to participate).
On the day of the workshop (October 1st 2012, at MICCAI), participants were provided with a new Test2 Set and organizers evaluated the results and presented them during the conference. After the workshop, a collaborative paper describing the data, the participant methods and summarizing the results will be prepared for submission to a relevant journal (Medical Image Analysis or IEEE Transactions on Medical Imaging). The list of accepted papers is available here.
Last update on October 7th: Challenge results and photos available!
How to download the data after the challenge ?
Please see here.
We encourage the submission of manuscripts to the 7th International Conference on Functional Imaging and Modeling of the Heart (FIMH 2013, London, UK).
This event is supported by our main sponsor: Toshiba Medical Visualization Systems
and by Pie Medical Imaging.
 J. Caudron, J. Fares, V. Lefebvre, P.-H. Vivier, C. Petitjean and J.-N. Dacher, Cardiac MR Assessment of Right Ventricular Function in Acquired Heart Disease: Factors of Variability, Academic Radiology, Volume 19, Issue 8, pp. 991-1002, 2012
 C. Petitjean and J.-N. Dacher, A review of segmentation methods in cardiac MRI, MedIA, 15(169-184), 2011