Journal of Pathology Informatics

ORIGINAL ARTICLE
Year
: 2020  |  Volume : 11  |  Issue : 1  |  Page : 10-

EpithNet: Deep regression for epithelium segmentation in cervical histology images


Sudhir Sornapudi1, Jason Hagerty2, R Joe Stanley1, William V Stoecker3, Rodney Long3, Sameer Antani3, George Thoma3, Rosemary Zuna4, Shellaine R Frazier5 
1 Department of Electrical and Computer Engineering, Missouri University of Science and Technology, Rolla, MO, USA
2 Department of Electrical and Computer Engineering, Missouri University of Science and Technology; Stoecker and Associates, Rolla, MO, USA
3 Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, DHHS, Bethesda, MD, USA
4 Department of Pathology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA
5 Department of Surgical Pathology, University of Missouri Hospitals and Clinics, Columbia, MO, USA

Correspondence Address:
Dr. R Joe Stanley
Department of Electrical and Computer Engineering, Missouri University of Science and Technology, 127 Emerson Electric Co. Hall, 301 W. 16th Street, Rolla, MO 65409-0040
USA

Background: Automated pathology techniques for detecting cervical cancer at the premalignant stage have advantages for women in areas with limited medical resources. Methods: This article presents EpithNet, a deep learning approach for the critical step of automated epithelium segmentation in digitized cervical histology images. EpithNet employs three regression networks of varying dimensions of image input blocks (patches) surrounding a given pixel, with all blocks at a fixed resolution, using varying network depth. Results: The proposed model was evaluated on 311 digitized histology epithelial images and the results indicate that the technique maximizes region-based information to improve pixel-wise probability estimates. EpithNet-mc model, formed by intermediate concatenation of the convolutional layers of the three models, was observed to achieve 94% Jaccard index (intersection over union) which is 26.4% higher than the benchmark model. Conclusions: EpithNet yields better epithelial segmentation results than state-of-the-art benchmark methods.


How to cite this article:
Sornapudi S, Hagerty J, Stanley R J, Stoecker WV, Long R, Antani S, Thoma G, Zuna R, Frazier SR. EpithNet: Deep regression for epithelium segmentation in cervical histology images.J Pathol Inform 2020;11:10-10


How to cite this URL:
Sornapudi S, Hagerty J, Stanley R J, Stoecker WV, Long R, Antani S, Thoma G, Zuna R, Frazier SR. EpithNet: Deep regression for epithelium segmentation in cervical histology images. J Pathol Inform [serial online] 2020 [cited 2021 Aug 1 ];11:10-10
Available from: https://www.jpathinformatics.org/article.asp?issn=2153-3539;year=2020;volume=11;issue=1;spage=10;epage=10;aulast=Sornapudi;type=0