Share Email Print
cover

Proceedings Paper

Prostate malignancy grading using gland-related shape descriptors
Author(s): Ulf-Dietrich Braumann; Patrick Scheibe; Markus Loeffler; Glen Kristiansen; Nicolas Wernert
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

A proof-of-principle study was accomplished assessing the descriptive potential of two simple geometric measures (shape descriptors) applied to sets of segmented glands within images of 125 prostate cancer tissue sections. Respective measures addressing glandular shapes were (i) inverse solidity and (ii) inverse compactness. Using a classifier based on logistic regression, Gleason grades 3 and 4/5 could be differentiated with an accuracy of approx. 95%. Results suggest not only good discriminatory properties, but also robustness against gland segmentation variations. False classifications in part were caused by inadvertent Gleason grade assignments, as a-posteriori re-inspections had turned out.

Paper Details

Date Published: 20 March 2014
PDF: 13 pages
Proc. SPIE 9041, Medical Imaging 2014: Digital Pathology, 90410M (20 March 2014); doi: 10.1117/12.2043225
Show Author Affiliations
Ulf-Dietrich Braumann, Univ. Leipzig (Germany)
Hochschule für Technik, Wirtschaft und Kultur Leipzig (Germany)
Patrick Scheibe, Univ. Leipzig (Germany)
Markus Loeffler, Univ. Leipzig (Germany)
Glen Kristiansen, Univ. Bonn (Germany)
Nicolas Wernert, Univ. Bonn (Germany)


Published in SPIE Proceedings Vol. 9041:
Medical Imaging 2014: Digital Pathology
Metin N. Gurcan; Anant Madabhushi, Editor(s)

© SPIE. Terms of Use
Back to Top