RT journal article
T1 Automatic segmentation of whole-slide H&E stained breast histopathology images using a deep convolutional neural network architecture
A1 Priego Torres, Blanca María
A1 Sánchez Morillo, Daniel
A1 Fernández Granero, Miguel Ángel
A1 García-Rojo, Marcial
A2 Anatomía PatológicaBiología Celular, Histología, Historia de la Ciencia, Medicina Legal y Forense y Toxicología
A2 Ingeniería en AutomáticaElectrónica, Arquitectura y Redes de Computadores
K1 Breast cancer
K1 Deep learning
K1 H&E staining
K1 Segmentation
K1 Whole-Slide Imaging
AB In this research, we propose a processing pipeline for the automatic segmentation of stained BC images presenting different types of histopathological patterns. Experimental results on a collection of patches of breast cancer images demonstrate how the designed processing pipeline performs properly regardless of the size, texture or any other colour-shape features typical of the malignant carcinomas considered in this study. The estimated segmentation accuracy and frequency-weighted intersection over union ( FWIoU ) were 95.62%, 92.52%, respectively. Additionally, a web-based platform which includes a slide-viewer and an annotation tool was developed. The automatic segmentation method proposed in this work was integrated into this platform and currently, it is being used as a decision-support tool by pathologists.
PB PERGAMON-ELSEVIER SCIENCE LTD
SN 0957-4174
YR 2020
FD 2020-03-18
LK http://hdl.handle.net/10498/30087
UL http://hdl.handle.net/10498/30087
LA eng
NO This version of the article was accepted for publication, after peer review and does not reflect post-acceptance improvements, or any corrections. The published version is available online (2020-03-18) at: https://doi.org/10.1016/j.eswa.2020.113387.
DS Repositorio Institucional de la Universidad de Cádiz
RD 10-may-2026