Will geometry and intensity properties be similar for all datasets? Yes, the field of view and image intensity (no contrast-enhanced scans will be used) will be similar for all datasets, including the training, testing and contest data. However, some typical individual variability is to be expected, e.g. anatomical differences, image artifacts due to dental fillings, etc. All data used in the contest represent a subset of a real clinical patient population. How will the results be evaluated? The evaluation will be done slice-wise using axial views, which better corresponds to the clinical contouring workflow. The two main criteria for evaluation will be the Hausdorff distance and volumetric overlap (Dice metric) between the reference expert and the tested automated 2-D binary masks. More details on the evaluation criteria can be found here. A Windows executable to test the results against the manual expert delineations provided with the training data can be downloaded here. How must the results be submitted? The results have to be submitted in the form of binary masks represented by unsigned char data, analogously to the manual annotations supplied with the training data (see Download/Submit section). The participants should use the University Health Network (UHN) File Portal system to submit their results: https://fileportal.uhn.on.ca. Please enter for the email address: stephane.allaire@rmp.uhn.on.ca and then proceed with the upload as advised. Please use the team or contact person’s name as part of the file name to uniquely identify your submission. What is the paper submission format? The Grand Challenge workshop will be using the same format as the MICCAI 2010 conference. The maximum number of pages is 8. After submitting your results, we will send you several automatically generated tables and figures, which you must include in the Results section of your paper. Please email the paper as a PDF file to vladimir.pekar@philips.com. The content of this website is copyrighted © 2010 Vladimir Pekar, John Kim, Stéphane Allaire, Arish Qazi, David A. Jaffray