The Cancer Digital Slide Archive

Developed in collaboration with Professor David Gutman, the Cancer Digital Slide Archive (CDSA) is an online resource providing access to 18,000+ whole-slide images of tumors from The National Cancer Institute's Cancer Genome Atlas (TCGA).

The CDSA is a flexible framework for data integration that puts whole-slide images, pathology reports, clinical/treatment variables and genomic annotations at the fingertips of users. Requiring only a web-browser, the CDSA lowers the barrier for users to access to TCGA pathology and clincial/genomic metadata. An integrated radiology viewer is also provided for glioblastoma tumors, enabling simultaneous review of pathology and radiology imaging. The TCGA glioblastomas are provided as a fully integrated example with more to follow.

DA Gutman, J Cobb, D Somanna, Y Park, F Wang, T Kurc, JH Saltz, DJ Brat, LAD Cooper, "Cancer Digital Slide Archive: an informatics resource to support integrated in silico analysis of TCGA pathology data," Journal of the American Medical Informatics Association JAMIA, In Press

Matlab Tools for Whole Slide Image Analysis

We have implemented a Matlab wrapper for Openslide to provide access to image content stored in propriety whole-slide image formats from within the Matlab programming environment. This library uses MEX/C++ to interface with the Openslide libraries and enables users to query slide properties and to read subregions of whole slide images into the Matlab workspace.

Two Point Correlation Functions for Tissue Analysis

The TPCF library implements the two-point correlation function (TPCF), an image feature for higher-order histology segmentation. The TPCF libraries enable users to develop statistical geometric signatures of tissues for image segmentation or classification tasks, using a fast convolution updating algorithm that performs the minimum calculations needed for exact results. The libraries are implemented in C/C++ and include both serial and Message Passing Interface versions for parallel calculation of dense features in very large images.

LAD Cooper, JH Saltz, U Catalyurek, K Huang, Acceleration of Two Point Correlation Function Calculation for Pathology Image Segmentation, First International Conference on Healthcare Informatics, Imaging and Systems Biology (HISB), pp. 174-81, San Jose, CA, July 2011 [ieee]

LAD Cooper, JH Saltz, R Machiraju, K Huang, Two-Point Correlation as a Feature for Histology Images: Feature Space Structure and Correlation Updating, Mathematical Methods in Bioimage Analysis Workshop, 23rd IEEE Conference on Computer Vision and Pattern Recognition, pp.79-86, San Francisco, CA, June 2010 [ieee]