Cancer Imaging Phenomics Toolkit (CaPTk)  1.7.0
Brain Cancer: Pseudoprogression Infiltration Index

This application provides an estimate of the pseudo-progression after radiotherapy in glioblastoma patients, via multi-parametric MRI analysis, as shown in [1].

REQUIREMENTS:

  1. Co-registered Multimodal MRI: T1, T1-Gd, T2, T2-FLAIR, DSC-MRI, DTI-AX, DTI-FA, DTI-RAD, DTI-TR. Ensure that these are the identified modalities in the drop-down menus next to each loaded image.
  2. Segmentation label of the demarcated region of interest (Label=1) in a single NIfTI (.nii.gz) file.
  3. The data for each patient should be organized in the following directory structure.
    • SubjectID
      1. CONVENTIONAL
        • "my_T1_file.nii.gz"
        • "my_T2_file.nii.gz"
        • "my_T1CE_file.nii.gz"
        • "my_FLAIR_file.nii.gz"
      2. DTI
        • "my_AX_file.nii.gz"
        • "my_FA_file.nii.gz"
        • "my_RAD_file.nii.gz"
        • "my_TR_file.nii.gz"
      3. PERFUSION
        • "my_PERFUSION_file.nii.gz"
      4. SEGMENTATION
        • "my_segmentation_file.nii.gz"
  4. The data of multiple patients should be organized in the above mentioned structure and reside under the same folder, e.g.:
    • Data_of_multiple_patients
      1. Subject_ID1
      2. Subject_ID2
      3. ...
      4. Subject_IDn

USAGE:

  • Pseudoprogression assessment on a batch of subjects.
    • "Train a new model":
      1. Select the "Training Directory", e.g., Data_of_multiple_patients (confirm the data follows the structure instructed above).
      2. Select the "Output Directory" where the trained model should be saved.
      3. Click on 'Confirm'.
      4. A pop-up window will confirm the completion of model training (~1.5*NoOfSubjects minutes).
      5. This application is also available as with a stand-alone CLI for data analysts to build pipelines around.
      6. NOTE: in the sample data, we are providing multiple duplicates of 5 unique subjects to show the training functionality at work; the model generated using these should NOT be used to generate results. System should have atleast 32GB RAM for calculating perfusion derivatives.
        PseudoProgressionEstimator.exe -t 0 -i C:/PseudoprogressionSubjects -o C:/PseudoprogressionModel
        
    • "Use existing model":
      1. Select the "Model Directory". Note that a model trained on a cohort of HUP can be found in ftp://www.nitrc.org/home/groups/captk/downloads/models/pseudoprogression.zip
      2. Select the "Test Directory", e.g., Data_of_multiple_patients (confirm the data follows the structure instructed above).
      3. Select the "Output Directory", where the user wants to save the output. The first and the second column of .csv will be distancce of sample from the hyperplance of pseudoprogression model and true recurrence model, respectively.
      4. Click on 'Confirm'.
      5. A pop-up window will confirm the completion of assessment of pseudoprogression (~1.5*NoOfSubjects minutes).
      6. This application is also available as with a stand-alone CLI for data analysts to build pipelines around:
        PseudoProgressionEstimator.exe -t 1 -i C:/PseudoprogressionSubjects -o C:/PseudoprogressionOutput -m C:/PseudoprogressionModel
        

RESULT INTERPRETATION:

  • 1st column: distance to hyperplane that classifies pseudo-progression versus the rest
  • 2nd column: distance to hyperplane that classifies true-progression versus the rest



Reference:

  1. H.Akbari, S.Bakas, M.Martinez-Lage, M.Nasrallah, M.Rozycki, S.Rathore, G.Shukla, S.Mohan, M.Bilello, C.Davatzikos, "Quantitative radiomics and machine learning to distinguish true progression from pseudoprogression in patients with GBM", ASNR 56th Annual Meeting, 2018.