Cancer Imaging Phenomics Toolkit (CaPTk)  1.9.0
Brain Cancer: Confetti

This is a method for automated extraction of white matter tracts of interest in a consistent and comparable manner over a large group of subjects without drawing the inclusion and exclusion ROIs, facilitating an easy correspondence between different subjects, as well as providing a representation that is robust to edema, mass effect, and tract infiltration [1-3]. Confetti includes three main steps:

  1. Connectivity signature generation for fibers
  2. Clustering of fibers using a mixture of multinomials (MNs) clustering method and Expectation-Maximization (EM) optimization framework
  3. Extraction of predefined white matter tracts.

REQUIREMENTS:

  1. Fiber set (Streamline) to be clustered: The fiber set can be generated using any tractography model, but the file should be saved in .Bfloat format (i.e. fiber format of Camino package). Different converters can be be used to convert .trk to .Bfloat and vice-versa.
  2. Track Density Images (TDI): When using GUI, it needs to be generated in the manner explained below; this constraint is not present when using Confetti via the command line.
  3. Parcellation of the brain into 87 Desikan/Freesurfer gray matter (GM) regions [4]

Generation of TDI Images with GUI:

  1. Freesurfer [4] is used with the Desikan atlas [5] to define 87 gray matter ROIs in the user diffusion space.
  2. Region IDs of the ROIs as used by Freesurfer is provided in the example file "{CaPTk_Sample_Data}/Confetti/input/freesurfer_ROIs.csv". (Note that the CSV file should be in ASCII format)
  3. TDIs must be generated using the probtrackx utility of FSL package [6] with its default parameters and 5000 seeds per voxel.
  4. Each TDI image is a whole brain voxel-map, with each voxel having the number of fibers passing through this voxel and reaching to one of the 87 gray matter ROIs defined by Freesurfer.
  5. In total, you should have 87 TDI, each corresponding to one ROI.

USAGE:

  1. Open Confetti UI using the 'Applications -> Confetti' menu option.
  2. Load the required images using "Streamline File" and "TDI Directory".
  3. Specify the output directory and click on 'Run Confetti'.
  4. Visually review the output tracks by double clicking respective fields on the populated list view.
  • This application is also available as with a stand-alone CLI for data analysts to build pipelines around, using the following example commands:
  1. Generation of connectivity signatures of the fibers that will be clustered into bundles: 
    ${CaPTk_InstallDir}/bin/Confetti signature -i tdi_paths_freesurfer_87ROIs.csv -f fibers.Bfloat -o signatures.csv
  2. Clustering of the generated fibers into bundles: 
    ${CaPTk_InstallDir}/bin/Confetti cluster -s signatures.csv -k 200 -o clusterIDs.csv
  3. Identification of specific tracts (requires an annotated example): 
    ${CaPTk_InstallDir}/bin/Confetti extract -t template/ -f fibers.Bfloat -c clusterIDs.csv -o tracts_

References:

  1. B.Tunc, M.Ingalhalikar, W.A.Parker, J.Lecoeur, N.Singh, R.L.Wolf, L.Macyszyn, S.Brem, R.Verma, "Individualized Map of White Matter Pathways: Connectivity-based Paradigm for Neurosurgical Planning", Neurosurgery. 79(4):568-77, 2016, DOI:10.1227/NEU.0000000000001183.
  2. B.Tunc, W.A.Parker, M.Ingalhalikar, R.Verma, "Automated tract extraction via atlas based Adaptive Clustering", NeuroImage. 102(2):596-607, 2014, DOI:10.1016/j.neuroimage.2014.08.021
  3. B.Tunc, A.R.Smith, D.Wasserman, X.Pennec, W.M.Wells, R.Verma, K.M.Pohl, "Multinomial Probabilistic Fiber Representation for Connectivity Driven Clustering", Inf Process Med Imaging. 23:730-41, 2013.
  4. B.Fischl, M.I.Sereno, A.M.Dale, "Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system", NeuroImage. 9:195-207, 1999, DOI:10.1006/nimg.1998.0396
  5. R.S.Desikan, F.Segonne, B.Fischl, B.Quinn, B.Dickerson, D.Blacker, R.Buckner, A.Dale, R.Maguire, B.Hyman, M.Albert, R.Killiany, "An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest", NeuroImage. 31(3):968-80, 2006, DOI:10.1016/j.neuroimage.2006.01.021
  6. M.Jenkinson, C.F.Beckmann, T.E.J.Behrens, M.W.Woolrich, S.M.Smith, "FSL", Neuroimage. 62(2):782-790, 2012, DOI:10.1016/j.neuroimage.2011.09.015