Software
CBIG is committed to advancing research by making publicly available datasets and software analytics acquired and/or developed by on-going research studies. We provide access to software developed by CBIG, which is available for public release.
Laboratory for Individualized Breast Radiodensity Assessment (LIBRA)
LIBRA 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") mammography images from a variety of mammography vendors. 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.
For more information and to download the LIBRA software, please visit the LIBRA project page.
Breast density estimation from digital mammography using the LIBRA algorithm developed by our group, based on an adaptive fuzzy C-means clustering dense tissue segmentation.
Cancer Imaging Phenomics Toolkit (CaPTk)
CaPTk is a software platform for analysis of radiographic cancer images, currently focusing on brain, breast, and lung cancer. CaPTk integrates advanced, validated tools performing various aspects of medical image analysis, including image segmentation, extensive panels of quantitative radiomic imaging features, and predictive modeling machine learning applications. CaPTk aims to facilitate the swift translation of advanced computational algorithms into routine clinical quantification, analysis, decision making, and reporting workflow. Its long-term goal is providing widely used technology that leverages the value of advanced imaging analytics in cancer prediction, diagnosis and prognosis, as well as in better understanding the biological mechanisms of cancer development.
For more information and to download the CaPTk software, please visit the CaPTk project page.
TomoLIBRA: Deep Learning for Volumetric Breast Density Estimation Using 3D DBT Reconstructed Slices
TomoLIBRA is the first deep learning model that can estimate volumetric breast density (VBD) and absolute dense volume (ADV) from 3D digital breast tomosynthesis (DBT) reconstructed slices. Using the GaNDLF framework, we trained a convolutional neural network (CNN) that can perform dense breast tissue segmentation on 3D DBT reconstructed slices, which are more commonly archived in clinical centers than the “raw” or “for processing” images. We envision that this model can be used to perform large retrospective breast density assessments and eventually perform prospective breast cancer risk assessments.
For more information and to download TomoLIBRA software, please visit the TomoLIBRA project page.
