Name
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. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1117/12.2043225 | Proceedings of SPIE |
Keywords | Field | DocType |
prostate cancer grading,Gleason grading,interrater disagreement,intrarater disagreement,prostate gland shape analysis,inverse solidity,inverse compactness,logistic regression model | Computer vision,Grading (education),Pattern recognition,Segmentation,Malignancy,Prostate,Artificial intelligence,Prostate cancer,Classifier (linguistics),Logistic regression,Shape analysis (digital geometry),Physics | Conference |
Volume | ISSN | Citations |
9041 | 0277-786X | 0 |
PageRank | References | Authors |
0.34 | 6 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ulf-dietrich Braumann | 1 | 66 | 14.28 |
| Patrick Scheibe | 2 | 1 | 2.72 |
| Markus Loeffler | 3 | 72 | 11.11 |
| Glen Kristiansen | 4 | 31 | 2.13 |
| Nicolas Wernert | 5 | 0 | 0.68 |