Share Email Print
cover

Journal of Medical Imaging

Evaluating stability of histomorphometric features across scanner and staining variations: prostate cancer diagnosis from whole slide images
Author(s): Patrick Leo; George Lee; Natalie N. C. Shih; Robin Elliott; Michael D. Feldman; Anant Madabhushi
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Quantitative histomorphometry (QH) is the process of computerized feature extraction from digitized tissue slide images to predict disease presence, behavior, and outcome. Feature stability between sites may be compromised by laboratory-specific variables including dye batch, slice thickness, and the whole slide scanner used. We present two new measures, preparation-induced instability score and latent instability score, to quantify feature instability across and within datasets. In a use case involving prostate cancer, we examined QH features which may detect cancer on whole slide images. Using our method, we found that five feature families (graph, shape, co-occurring gland tensor, sub-graph, and texture) were different between datasets in 19.7% to 48.6% of comparisons while the values expected without site variation were 4.2% to 4.6%. Color normalizing all images to a template did not reduce instability. Scanning the same 34 slides on three scanners demonstrated that Haralick features were most substantively affected by scanner variation, being unstable in 62% of comparisons. We found that unstable feature families performed significantly worse in inter- than intrasite classification. Our results appear to suggest QH features should be evaluated across sites to assess robustness, and class discriminability alone should not represent the benchmark for digital pathology feature selection.

Paper Details

Date Published: 24 October 2016
PDF: 11 pages
J. Med. Imag. 3(4) 047502 doi: 10.1117/1.JMI.3.4.047502
Published in: Journal of Medical Imaging Volume 3, Issue 4
Show Author Affiliations
Patrick Leo, Case Western Reserve Univ. (United States)
George Lee, Case Western Reserve Univ. (United States)
Natalie N. C. Shih, Univ. of Pennsylvania (United States)
Robin Elliott, Case Western Reserve Univ. (United States)
Michael D. Feldman, The Univ. of Pennsylvania (United States)
Anant Madabhushi, Case Western Reserve Univ. (United States)


© SPIE. Terms of Use
Back to Top