1. Prostate malignancy grading using gland-related shape descriptors
- Author
-
Ulf-Dietrich Braumann, Patrick Scheibe, Nicolas Wernert, Markus Loeffler, and Glen Kristiansen
- Subjects
business.industry ,Pattern recognition ,Logistic regression ,Gleason grade ,medicine.disease ,Malignancy grading ,Prostate cancer ,medicine.anatomical_structure ,Prostate ,medicine ,Segmentation ,Artificial intelligence ,business ,Classifier (UML) ,Mathematics ,Shape analysis (digital geometry) - 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.
- Published
- 2014
- Full Text
- View/download PDF