Laboratory for Individualized Breast Radiodensity Assessment
The Laboratory for Individualized Breast Radiodensity Assessment (LIBRA), a software package developed by the Computational Breast Imaging Group (CBIG) at the University of Pennsylvania, is a fully-automatic breast density estimation software solution based on a published algorithm that works on either raw (i.e., “FOR PROCESSING”) or vendor post-processed (i.e., “FOR PRESENTATION”) digital mammography images from two vendors: GE Healthcare and Hologic. Based on an algorithm published in Medical Physics, LIBRA has been applied to over 30,000 screening exams and is being increasingly utilized in larger studies.
LIBRA has two modes of operations:
- An easy-to-use interactive mode with Graphical-User-Interface where the user is prompted to select either a single DICOM image or a folder of DICOM images, an output folder for the results, and whether they wish to save intermediate files.
- A command-line interface amenable to batch processing and scripting, where the user can explicitly define the input and output paths.
The amount of fibroglandular tissue content in the breast as estimated mammographically, commonly referred to as breast percent density (PD%), is one of the most significant risk factors for developing breast cancer. Approaches to quantify breast density commonly focus on either semiautomated methods or visual assessment, both of which are highly subjective. This software package was developed to be a fully-automated density estimation method that works on both raw (i.e., “FOR PROCESSING”) and vendor postprocessed (i.e., “FOR PRESENTATION”) digital mammography images,and has thus far been validated to work on GE Healthcare and Hologic digital mammography systems [*].
Briefly, the software first applies an edge-detection algorithm to delineate the boundary of the breast and the boundary of the pectoral muscle. Following the segmentation of the breast, an adaptive multi-class fuzzy c-means algorithm is applied to identify and partition the mammographic breast tissue area, into multiple regions (i.e., clusters) of similar intensity. These clusters are then aggregated by a support-vector machine classifier to a final dense tissue area, segmentation. The ratio of the segmented absolute dense area to the total breast area is then used to obtain a measure of breast percent density (PD%).
The software generates both quantitative estimates of breast area, dense area and PD% that are stored in a comma separated text file (.csv, openable by Excel) as well as a .JPG image of the breast and density segmentations overlaid on a window-levelled version of the mammogram amenable for publication to a user-defined directory, in addition to several optional files, as described in the Manual Section.
*DISCLAIMER: Density estimation on mammograms from other vendors has not been validated, therefore the performance and the quality of segmentation is not guaranteed. However the breast segmentation algorithm within LIBRA generally works well across all vendors and thus may be of general use in a research context.
The LIBRA software is freely available under a BSD-style open source license that is compatible with the Open Source Definition by The Open Source Initiative and contains no restrictions on use of the software. The full license text is included with the distribution package and available online.
LIBRA Manual: PDF version of software manual.
The LIBRA software package is primarily distributed as a pre-compiled, stand-alone executable to ensure platform-wide compatibility; the MATLAB source files are also provided for development purposes. Currently, the stand-alone version of LIBRA is available for 64-bit versions of Windows, with solutions for Linux and Mac OS being actively worked on. Running the stand-alone version will require downloading and installing MATLAB Compiler Runtime version R2013A (8.1), available for download at MathWorks, Inc. at the link below. Using the MATLAB source files (for any Operating System) will require MATLAB main license and access to certain toolboxes, listed below with links to the MathWorks, Inc. product pages describing them.
Prerequisites - Stand-alone executable
Operating System Requirements
Windows 7, 64-bit. Note: other 64-bit versions of Windows (i.e., XP, Vista and 8) may also be compatible, but have not been thoroughly evaluated at this time.
|MCR (Version 8.1)||R2013A 64-bit||MATLAB Compiler Runtime enables program to run without MATLAB installed.|
Prerequisites - MATLAB source files
Operating System Requirements
Windows 32-bit/64-bit, Mac OS 32-bit/64-bit, Linux 32-bit/64-bit
MATLAB Version Requirements
MATLAB 2012A, 2012B, 2013A, 2014A, and 2014B 32-bit/64-bit. Note: other versions of MATLAB may be compatible, but have not been thoroughly evaluated at this time. CAUTION: There is a known bug and incompatibility to MATLAB R2013b (version: 22.214.171.1245) on Windows that the dicomread.m function would randomly crash and terminate the matlab routine.
|Curve Fitting Toolbox||MathWorks, Inc. MATLAB Curve Fitting Toolbox|
|Image Processing Toolbox||MathWorks, Inc. MATLAB Image Processing Toolbox|
|Signal Processing Toolbox||MathWorks, Inc. MATLAB Signal Processing Toolbox|
|Statistics Toolbox||MathWorks, Inc. MATLAB Statistics Toolbox|
|Symbolic Math Toolbox||MathWorks, Inc. MATLAB Symbolic Math Toolbox|