Segmentation of breast cancer histopathology tissue micro-arrays for computer-aided diagnosis in pathology


Computer-Aided Diagnosis (CAD) systems for pathologists can act as an intelligent digital assistant supporting automated grading and morphometric-based discovery of tissue features that are important in cancer diagnosis and patient prognosis. Automated image segmentation is an essential component of computer-based grading in CAD. We describe a novel tissue segmentation algorithm using local feature-based multiphase active contours in a globally convex formulation. By decomposing the image into intensity and chromaticity channels we utilize multiple feature based fitting terms to drive the active contour evolution for effective stromal-epithelial separation. Experimental results using the Stanford Tissue MicroArray (TMA) database shows promising stromal/epithelial superpixel segmentation.

Some Results [1]


(a) Input (b) Groundtruth (c) Intensity feature (d) Chromaticity feature (e) Our scheme[1] (f) Contours

"Histopathology - ἱστόπάθολογί - Histo-tissue, patho-disease"

More Results:

Please see our poster at Figshare!


[1] V. B. S. Prasath, F. Bunyak, P. Dale, S. R. Frazier, K. Palaniappan. Segmentation of breast cancer tissue microarrays for computer-aided diagnosis in pathology. First IEEE Healthcare Technology Conference: Translational Engineering in Health & Medicine (IEEE HIC 2012), Houston, TX, USA.

[2] V. B. S. Prasath, F. Bunyak, P. Dale, S. R. Frazier, K. Palaniappan. Stromal-epithelial separation for breast cancer tissue microarrays histopathology. Poster at the Missouri Life Sciences Week 2014. figshare: 10.6084/m9.figshare.997514.

In preparation:

Segmentation of breast cancer tissue microarrays for stromal-epithelial separation in pathology.

State of the art in digital histopathology image processing: Breast cancer.

Back to Histopathology Project Main page. Back to Research Page.