Implementation of STAPLE Variants

We are distributing the ITK implementation of several of our contributions to the STAPLE algorithm:

  • the continuous STAPLE algorithm [1], derived for the study of diffusion tensor images as well as scalar images, is made available. More details can be found here.
  • the MAP STAPLE algorithm [2], whose goal is to include priors on the performance parameters estimation, particularly useful when some labels in the image could not be delineated by one or several raters. More details in the related publication: [2].
  • Soon, we will also be releasing the implementation of local MAP STAPLE [3] and the estimation of inferential uncertainty on the estimated performance parameters from STAPLE [4]. This will be made available from the same source.

Downloading and Licensing

The latest version of the package for all STAPLE variants and its base components are available through the same website as the original STAPLE algorithm:

A usage tutorial on STAPLE is also available from this website.

These softwares and their base components are distributed under the OSI Open Software License (OSL-3.0). Please refer to the following link for more explanations. If you're using these softwares for publications, please cite the matching paper(s) as a reference [1,2,3,4].


  1. A Continuous STAPLE for Scalar, Vector and Tensor Images: An Application to DTI Analysis
    O. Commowick and S. K. Warfield.
    IEEE Transactions on Medical Imaging, 28(6):838-846, June 2009. pdf doi
  2. Incorporating Priors On Expert Performance Parameters for Segmentation Validation and Label Fusion: a Maximum A Posteriori STAPLE
    O. Commowick and S. K. Warfield.
    In Proceedings of the 13th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI'10), Part III, volume 6363 of LNCS, pages 25-32, September 2010. pdf
  3. Estimating A Reference Standard Segmentation with Spatially Varying Performance Parameters: Local MAP STAPLE
    O. Commowick, A. Akhondi-Asl and S. K. Warfield.
    IEEE Transactions on Medical Imaging, 31(8):1593-1606, August 2012. pdf doi
  4. Estimation of Inferential Uncertainty in Assessing Expert Segmentation Performance from STAPLE
    O. Commowick and S. K. Warfield.
    IEEE Transactions on Medical Imaging, 29(3):771-780, March 2010. pdf doi