Medical College of Wisconsin
CTSICores SearchResearch InformaticsREDCap

Accurate segmentation of prostate cancer histomorphometric features using a weakly supervised convolutional neural network. J Med Imaging (Bellingham) 2020 Sep;7(5):057501

Date

10/17/2020

Pubmed ID

33062803

Pubmed Central ID

PMC7550797

DOI

10.1117/1.JMI.7.5.057501

Scopus ID

2-s2.0-85096654686 (requires institutional sign-in at Scopus site)   6 Citations

Abstract

Purpose: Prostate cancer primarily arises from the glandular epithelium. Histomophometric techniques have been used to assess the glandular epithelium in automated detection and classification pipelines; however, they are often rigid in their implementation, and their performance suffers on large datasets where variation in staining, imaging, and preparation is difficult to control. The purpose of this study is to quantify performance of a pixelwise segmentation algorithm that was trained using different combinations of weak and strong stroma, epithelium, and lumen labels in a prostate histology dataset. Approach: We have combined weakly labeled datasets generated using simple morphometric techniques and high-quality labeled datasets from human observers in prostate biopsy cores to train a convolutional neural network for use in whole mount prostate labeling pipelines. With trained networks, we characterize pixelwise segmentation of stromal, epithelium, and lumen (SEL) regions on both biopsy core and whole-mount H&E-stained tissue. Results: We provide evidence that by simply training a deep learning algorithm on weakly labeled data generated from rigid morphometric methods, we can improve the robustness of classification over the morphometric methods used to train the classifier. Conclusions: We show that not only does our approach of combining weak and strong labels for training the CNN improve qualitative SEL labeling within tissue but also the deep learning generated labels are superior for cancer classification in a higher-order algorithm over the morphometrically derived labels it was trained on.

Author List

Bukowy JD, Foss H, McGarry SD, Lowman AK, Hurrell SL, Iczkowski KA, Banerjee A, Bobholz SA, Barrington A, Dayton A, Unteriner J, Jacobsohn K, See WA, Nevalainen MT, Nencka AS, Ethridge T, Jarrard DF, LaViolette PS

Authors

Anjishnu Banerjee PhD Associate Professor in the Data Science Institute department at Medical College of Wisconsin
Samuel Bobholz PhD Assistant Professor in the Radiology department at Medical College of Wisconsin
Peter LaViolette PhD Professor in the Radiology department at Medical College of Wisconsin
Andrew S. Nencka PhD Director, Associate Professor in the Radiology department at Medical College of Wisconsin