Results of the Contouring Challenge

• Welcome
• 1. The Images
• 2. The Contenders
• 3. Contour Evaluation
• Experiment i
• Experiment ii
• Experiment iii
• Experiment iv
• Discussions

Please give feedback, to

or at the

of the

Experiment i

This experiment investigates the value of the phantom image for the purpose of evaluating contouring performance. A good phantom image is one that poses a similar challenge to the contouring method, as would a real (patient) image. The null-hypothesis is that both phantom and patient images lead to the same distribution of accuracy and variability scores across all contender VOIs.

Experimental design and results will be reported in the publication below, currently in press (30 May 2012)

For more information contact

T. Shepherd, M. Teräs, R. R. Beichel, R. Boellaard, M. Bruynooghe, V. Dicken, M. J. Gooding, P. J Julyan, J. A. Lee, S. Lefevre, M. Mix, V. Naranjo, X. Wu, H. Zaidi, Z. Zeng and H. Minn., "Comparative Study with New Accuracy Metrics for Target Volume Contouring in PET Image Guided Radiation Therapy", IEEE Transactions on Medical Imaging (TMI), in press (accepted May 2012)