Python implementation of the Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm for generating ground truth volumes from a set of binary segmentations.
The STAPLE algorithm is described in S. Warfield, K. Zou, W. Wells, Validation of image segmentation and expert quality with an expectation-maximization algorithm in MICCAI 2002: Fifth International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer-Verlag, Heidelberg, Germany, 2002, pp. 298-306.
$ pip install staple
$ staple seg_1.nii.gz seg_2.nii.gz seg_3.nii.gz result.nii.gz
- The SimpleITK implementation is about 16 times faster for the test images (0.7 s vs 11.8 s). The implementation in this repository is mostly for educational purposes.
- Markov random field (MRF) postprocessing is not implemented (nor is it in the ITK version). If you need STAPLE with MRF, check out NiftySeg.