Demo

Model parameters

Task: Team K2's solution for WMH segmentation grand challenge (http://wmh.isi.uu.nl/) at MICCAI 2017.
Imaging modality: MRI
Organ: Brain
Input(s): Brain Axial T1-W MRI and FLAIR
Output(s): White matter hyperintensities binary segmentation maps
Training set size: 60 / Test set size: 110
Performance: Dice: %77, H95: 9.79 mm
contact: Alireza Mehrtash, email: mehrtash at bwh.harvard.edu
Size: 6.4 GB

Demo

Command-line guide

docker pull deepinfer/wmh-segmenter-k2

Example

docker run -t -v ~/data/wmh_test/:/data deepinfer/whm-segmenter-k2\
                   --ModelName wmh_segmenter\
                   --InputT1Volume /data/T1.nrrd\
                   --InputFLAIRVolume /data/flair.nrrd\
                   --OutputLabel /data/wmh_label.nrrd \
                   --verbose

Inputs

[Mandatory]
ModelName: ('wmh_segmenter')
InputT1Volume: (an existing filename locating the T1-Weighted MRI of the brain)
InputFLAIRVolume: (an existing filename locating the FLAIR MRI of the brain)
OutputLabel: (output path of the white matter hyperintensity (WMH) label)

[Optional]
verbose : 
verbose mode for printing additional details about the procedure.

Reuse and Citations

The WMH-Segmenter-K2 model is licensed under Slicer License.
For attribution in academic contexts, please cite the following work(s):

Kuijf, Hugo J., et al. "Standardized assessment of automatic segmentation of white matter hyperintensities and results of the WMH segmentation challenge." 
IEEE transactions on medical imaging 38.11 (2019): 2556-2568.

BibTeX citation

@article{kuijf2019standardized,
  title={Standardized assessment of automatic segmentation of white matter hyperintensities and results of the WMH segmentation challenge},
  author={Kuijf, Hugo J and Biesbroek, J Matthijs and De Bresser, Jeroen and Heinen, Rutger and Andermatt, Simon and Bento, Mariana and Berseth, Matt and Belyaev, Mikhail and Cardoso, M Jorge and Casamitjana, Adria and others},
  journal={IEEE transactions on medical imaging},
  volume={38},
  number={11},
  pages={2556--2568},
  year={2019},
  publisher={IEEE}
}