Detection and classification of cancer in whole slide breast histopathology images using deep convolutional networks

•Commonly studied scenario considers only binary cancer vs. no cancer classification.•Our system classifies whole slide breast biopsies into five diagnostic categories.•Pipeline of fully convolutional networks localizes diagnostically relevant regions.•Convolutional neural network classifies detecte...

Full description

Saved in:
Bibliographic Details
Published in:Pattern recognition 2018-12, Vol.84, p.345-356
Main Authors: Gecer, Baris, Aksoy, Selim, Mercan, Ezgi, Shapiro, Linda G., Weaver, Donald L., Elmore, Joann G.
Format: Article
Language:English
Subjects:
Breast histopathology
Deep learning
Digital pathology
Multi-class classification
Region of interest detection
Saliency detection
Whole slide imaging
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Commonly studied scenario considers only binary cancer vs. no cancer classification.•Our system classifies whole slide breast biopsies into five diagnostic categories.•Pipeline of fully convolutional networks localizes diagnostically relevant regions.•Convolutional neural network classifies detected regions of interest in whole slides.•Experiments show that our method is compatible with predictions of 45 pathologists. Generalizability of algorithms for binary cancer vs. no cancer classification is unknown for clinically more significant multi-class scenarios where intermediate categories have different risk factors and treatment strategies. We present a system that classifies whole slide images (WSI) of breast biopsies into five diagnostic categories. First, a saliency detector that uses a pipeline of four fully convolutional networks, trained with samples from records of pathologists’ screenings, performs multi-scale localization of diagnostically relevant regions of interest in WSI. Then, a convolutional network, trained from consensus-derived reference samples, classifies image patches as non-proliferative or proliferative changes, atypical ductal hyperplasia, ductal carcinoma in situ, and invasive carcinoma. Finally, the saliency and classification maps are fused for pixel-wise labeling and slide-level categorization. Experiments using 240 WSI showed that both saliency detector and classifier networks performed better than competing algorithms, and the five-class slide-level accuracy of 55% was not statistically different from the predictions of 45 pathologists. We also present example visualizations of the learned representations for breast cancer diagnosis.
ISSN:0031-3203
1873-5142
DOI:10.1016/j.patcog.2018.07.022