Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe) captures subtle anisotropic differences in disease pathologies by measuring entropy of co-occurrences of voxel-level gradient orientations on imaging computed within a local neighborhood. CoLlAGe is based on the hypothesis that disruption in tissue microarchitecture can be quantified on imaging by measuring the disorder in voxel-wise gradient orientations. CoLlAGe involves assigning every image voxel a ‘disorder value’ associated with the co-occurrence matrix of gradient orientations computed around every voxel. After feature extraction, the subsequent distribution or different statistics such as mean, median, variance etc can be computed and used in conjunction with a machine learning classifier to distinguish similar appearing pathologies. The feasibility of CoLlAGe in distinguishing cancer from treatment confounders/benign conditions and characterizing molecular subtypes of cancers has been demonstrated in the context of multiple challenging clinical problems.

CoLlAGe quantifies intensity gradient differences that reflect underlying tissue heterogeneity, thus distinguishing similar appearing pathologies.

COLlAGe is available through the RadXTools ITCR project (1U01CA248226-01) as a Python module that can be integrated into major image informatics platforms (e.g. 3D Slicer) as well Docker/PIP.

If you use or publish research based on this implementation of COLlAGe, please cite:

Prasanna, P., Tiwari, P., & Madabhushi, A. (2016). Co-occurrence of Local Anisotropic Gradient Orientations (CoLlAGe): A new radiomics descriptor. Scientific Reports, volume 6, Article number: 37241 (2016).

COLlAGe has also been utilized in a variety of applications and diseases thus far by our group:

  • Nathaniel M. Braman, Maryam Etesami, Prateek Prasanna, Christina Dubchuk, Hannah Gilmore, Pallavi Tiwari, Donna Plecha, and Anant Madabhushi, “Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI”, Breast Cancer Research, 19:57 (2017).
  • Rakesh Shiradkar, Soumya Ghose, Ivan Jambor, Pekka Taimen, Otto Ettala, Andrei S. Purysko, and Anant Madabhushi, “Radiomic features from pretreatment biparametric MRI predict prostate cancer biochemical recurrence: Preliminary findings.”, J. Magn. Reson. Imaging, doi:10.1002/jmri.26178 (2018).
  • Niha Beig, Jay Patel, Prateek Prasanna, Virginia Hill, Amit Gupta, Ramon Correa, Kaustav Bera, Salendra Singh, Sasan Partovi, Vinay Varadan, Manmeet Ahluwalia, Anant Madabhushi, and Pallavi Tiwari, “Radiogenomic analysis of hypoxia pathway is predictive of overall survival in Glioblastoma”, Scientific Reports, volume 8, Article number: 7 (2018).