Deep Analysis of CNN Settings for New Cancer Whole-slide Histological Images Segmentation: The Case of Small Training Sets

Sonia Mejbri, Camille Franchet, Reshma Ismat-Ara, Josiane Mothe, Pierre Brousset, Emmanuel Faure

2019

Abstract

Accurate analysis and interpretation of stained biopsy images is a crucial step in the cancer diagnostic routine which is mainly done manually by expert pathologists. The recent progress of digital pathology gives us a challenging opportunity to automatically process these complex image data in order to retrieve essential information and to study tissue elements and structures. This paper addresses the task of tissue-level segmentation in intermediate resolution of histopathological breast cancer images. Firstly, we present a new medical dataset we developed which is composed of hematoxylin and eosin stained whole-slide images wherein all 7 tissues were labeled by hand and validated by expert pathologist. Then, with this unique dataset, we proposed an automatic end-to-end framework using deep neural network for tissue-level segmentation. Moreover, we provide a deep analysis of the framework settings that can be used in similar task by the scientific community.

Paper Citation

in Harvard Style

Mejbri S., Franchet C., Ismat-Ara R., Mothe J., Brousset P. and Faure E. (2019). Deep Analysis of CNN Settings for New Cancer Whole-slide Histological Images Segmentation: The Case of Small Training Sets. In Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2019) - Volume 2: BIOIMAGING; ISBN 978-989-758-353-7, SciTePress, pages 120-128. DOI: 10.5220/0007406601200128

in Bibtex Style

@conference{bioimaging19,
author={Sonia Mejbri and Camille Franchet and Reshma Ismat-Ara and Josiane Mothe and Pierre Brousset and Emmanuel Faure},
title={Deep Analysis of CNN Settings for New Cancer Whole-slide Histological Images Segmentation: The Case of Small Training Sets},
booktitle={Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2019) - Volume 2: BIOIMAGING},
year={2019},
pages={120-128},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007406601200128},
isbn={978-989-758-353-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2019) - Volume 2: BIOIMAGING
TI - Deep Analysis of CNN Settings for New Cancer Whole-slide Histological Images Segmentation: The Case of Small Training Sets
SN - 978-989-758-353-7
AU - Mejbri S.
AU - Franchet C.
AU - Ismat-Ara R.
AU - Mothe J.
AU - Brousset P.
AU - Faure E.
PY - 2019
SP - 120
EP - 128
DO - 10.5220/0007406601200128
PB - SciTePress