Cancer Imaging Phenomics Toolkit (CaPTk)  1.7.0
Related Pages
Here is a list of all related documentation pages:
[detail level 123]
 Installation Instructions
 Getting StartedThe following sections give a high-level overview of CaPTk, with the intention of familiarizing a new user with the platform functionalities and interface
 Keyboard ShortcutsThese are the keyboard shortcuts available on CaPTk:
 ComponentsCaPTk has been designed as a modular platform, currently incorporating components for:

  1. User interaction (e.g., coordinate definition, region annotation, and spherical approximation of abnormalities)
  2. Image pre-processing (e.g., smoothing, bias correction, co-registration, normalization)
  3. Image segmentation
  4. Feature extraction
  5. Specialized diagnostic analysis
 Supported ImagesCurrently, CaPTk supports visualization of MR, CT, PET, X-Ray and Full-field Digital Mammography (FFDM) images in both NIfTI (i.e., .nii, .nii.gz) and DICOM (i.e., .dcm) formats
 Image LoadingThe File -> Load menu is used to load all image types
 Image VisualizationSliders on each visualization panel (highlighted in green in the figure below) control the movement across respective axes
 Tab Docking (Windows-only)Double clicking on the tab bar will dock/undock the entire section, resulting in larger visualization panels in your monitor
 Coordinate definition (Seed-point initialization)The Seed Points tab includes two general types of initialization (i.e., tumor and tissue points), controlled by the respective radio buttons
 Label Annotation/Drawing PanelThis panel provides the ability to annotate Regions of Interest (ROIs) and save them as masks
 Pre-processingImage pre-processing is essential to quantitative image analysis
 Utilities (Command-line only)To make pipeline construction using CaPTk easier, a bunch of utilities have been provided
 SegmentationCurrently there are two ways to produce segmentation labels for various structures in the images that are loaded:

  1. using the Geodesic Distance Transform
  2. using ITK-SNAP
  3. Skull Stripping (Deep Learning based)
  4. Tumor Segmentation (Deep Learning based)
 Feature ExtractionThe feature extraction tab in CaPTk enables clinicians and other researchers to easily extract feature measurements, commonly used in image analysis, and conduct large-scale analyses in a repeatable manner
 Specialized ApplicationsVarious Specialized Applications are currently incorporated in CaPTk focusing on brain tumors, breast cancer, and lung nodules as shown in the figure below
 How To GuidesThis section provides step-by-step guidance to apply the CaPTk functionalities:
 Pre-processingContains the following applications:
 DICOM to NIfTI conversionThis tool converts a DICOM series into the Neuroimaging Informatics Technology Initiative (NIfTI) file format
 Image Co-registrationThis tool registers multiple moving images to a single target image using the Greedy Registration technique [1]
 Denoise-SUSAN (ITK filter)This tool smooths an image already loaded in CaPTk, to remove any high frequency intensity variations (i.e., noise), using the SUSAN algorithm [1]
 N4 Bias Correction (ITK filter)This tool corrects an image, already loaded in CaPTk, for magnetic field inhomogeneities using a non-parametric method [1]
 Histogram MatchingThis tool normalizes the intensity profile of an input image based on the intensity profile of a reference image [1]
 Z-Scoring NormalizationThis tool does a z-scoring based normalization on the loaded image
 Segmentation
 Feature ExtractionREQUIREMENTS:

  1. An image or a set of co-registered images
 Specialized Applications (SAs) UsageContains the following applications:
 Brain Cancer: WhiteStripe NormalizationThis algorithm normalizes conventional brain MRI scans [1] by detecting a latent subdistribution of normal tissue and linearly scaling the histogram of images
 Brain Cancer: Population AtlasThis application facilitates the users to generate population atlases for patients with different tumor subgroups [1] to emphasize their heterogeneity
 Brain Cancer: ConfettiThis 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]
 Brain Cancer: Glioblastoma Infiltration Index (Recurrence)This application provides a probability map of deeply infiltrating tumor in the peritumoral edema/invasion region that largely agrees with subsequent recurrence in de novo glioblastoma patients, via multi-parametric MRI analysis, as shown in [1-3]
 Brain Cancer: Pseudoprogression Infiltration IndexThis application provides an estimate of the pseudo-progression after radiotherapy in glioblastoma patients, via multi-parametric MRI analysis, as shown in [1]
 Brain Cancer: Glioblastoma Survival Prediction IndexThis application provides the survival prediction index (SPI) of de novo glioblastoma patients by using baseline pre-operative multi-parametric MRI analysis [1]
 Brain Cancer: Glioblastoma EGFRvIII Surrogate Index (PHI Estimator)This application evaluates the Epidermal Growth Factor Receptor splice variant III (EGFRvIII) status in primary glioblastoma, by quantitative pattern analysis of the spatial heterogeneity of peritumoral perfusion dynamics from Dynamic Susceptibility Contrast (DSC) MRI scans, through the Peritumoral Heterogeneity Index (PHI / φ-index) [1-3]
 Brain Cancer: Glioblastoma EGFRvIII SVM IndexThis application provides the detection of EGFRvIII mutation status of de novo glioblastoma patients by using baseline pre-operative multi-parametric MRI analysis [1]
 Brain Cancer: Glioblastoma Molecular Subtype PredictionThis application provides the prediction of the molecular subtype of de novo glioblastoma patients using baseline pre-operative multi-parametric MRI analysis [1]
 Breast Cancer: Breast Density Estimation (LIBRA)The Laboratory for Individualized Breast Radiodensity Assessment (LIBRA) is a software application for fully-automated breast density segmentation in full-field digital mammography (FFDM) [1]
 Breast Cancer: Texture Feature ExtractionThis application extracts the Texture Features as described in the paper [1]
 Breast Cancer: Breast SegmentationThis application uses LIBRA [1] to extract the breast region in the loaded image
 Lung Cancer: Radiomics Analysis of Lung CancerThis application provides a fully automatic segmentation of lung nodules and prediction of survival and nodal failure risks as a three step workflow[1]
 Brain Cancer: Directionality EstimatorThis application estimates the volumetric changes and their directionality for a given ROI across two timepoints [1] and also returns the projections of the boundary point with the maximum distance from a seedpoint (with label TU) in each of the 3 2D visualized planes
 Miscellaneous: Perfusion AlignmentThis application does perfusion alignment of the input Dynamic-Susceptibility Contrast-enhanced (DSC) MRI scan
 Miscellaneous: Perfusion DerivativesThis application extracts various measurements from a Dynamic-Susceptibility Contrast-enhanced (DSC) MRI scan
 Miscellaneous: Diffusion DerivativesThis application extracts various measurements from a Diffusion Weighted MRI scan
 Miscellaneous: Training ModuleThis applicatiparameterizeachine">Support Vector Machineparameterize
 Miscellaneous: PCA Volume ExtractionThis application extracts the principal components from DSC-MRI scans as mentioned in [1]
 Scientific Findings using CaPTkThis section presents examples of applications using CaPTk:
 Non-invasive Imaging Biomarker of EGFRvIII in Glioblastoma Patients
 Prediction of Overall Survival in Glioblastoma Patients
 Probability Maps of Potential Recurrence of Glioblastoma Tumors
 Imaging Biomarkers Related to Cancer Risk and Development of Breast Cancer
 Technical ReferenceThis section gives further technical details for all previous documentation material
 Further Application Details and AssumptionsContents
 Build CaPTk from SourceSource code for the CaPTk graphical interface and applications is distributed for sites that wish to examine the code, collaborate with CBICA in future development, and for compatibility
 For DevelopersContents
 Download InstructionsVisit our Download Page hosted in NIH-supported NITRC (https://www.nitrc.org/frs/?group_id=1059), to download the CaPTk source code and binaries
 Changelog: Release Notes
 People (Credits)