Breast Cancer: Texture Feature Extraction

This application extracts radiomic features related to breast cancer risk, as described in the paper [1]. It is based upon LIBRA for segmentation of the breast and the lattice-based mode of Feature Extraction backend to perform radiomic feature calculations.

REQUIREMENTS:

• Raw (i.e., "FOR PROCESSING") or vendor post-processed (i.e., "FOR PRESENTATION") FFDM images.
• FFDM vendors currently supported: Hologic and GE Healthcare.
• Each image stored in a separate folder (required only in GUI).

USAGE:

1. Load the FFDM image using the 'File -> Load -> Images' menu option.
2. Launch Breast Texture Feature Extraction using the 'Applications -> [Breast Cancer] Texture Pipeline' menu option.
3. Specify the output directory.

OUTPUT:

• output.csv: CSV file with the extracted radiomic features.
• featureMap/: Folder of feature maps, one for each radiomic feature, reflecting the spatial feature distributions within the breast.
  • This application is also available as with a stand-alone CLI for data analysts to build pipelines around, using the following example command:

    ${CaPTk_InstallDir}/bin/BreastTexturePipeline.exe -i C:/input/image.dcm -o C:/outputDir -d 1 
    

---

References:

1. Y.Zheng, B.M.Keller, S.Ray, Y.Wang, E.F.Conant, J.C.Gee, D.Kontos, "Parenchymal texture analysis in digital mammography: A fully automated pipeline for breast cancer risk assessment", Medical Physics, 2015.

---