A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images

Abstract The mitotic activity index is a key prognostic measure in tumour grading. Microscopy based detection of mitotic nuclei is a significant overhead and necessitates automation. This work proposes deep CNN based multi-phase mitosis detection framework “MP-MitDet” for mitotic nuclei identificati...

Full description

Bibliographic Details
Main Authors: Anabia Sohail, Asifullah Khan, Noorul Wahab, Aneela Zameer, Saranjam Khan
Format: Article
Language:English
Published: Nature Publishing Group 2021-03-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-021-85652-1
id doaj-0e4a3931d80c41a0a127dde5e87580bd
record_format Article
spelling doaj-0e4a3931d80c41a0a127dde5e87580bd2021-03-21T12:36:24ZengNature Publishing GroupScientific Reports2045-23222021-03-0111111810.1038/s41598-021-85652-1A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological imagesAnabia Sohail0Asifullah Khan1Noorul Wahab2Aneela Zameer3Saranjam Khan4Pattern Recognition Lab, DCIS, Pakistan Institute of Engineering and Applied Sciences (PIEAS)Pattern Recognition Lab, DCIS, Pakistan Institute of Engineering and Applied Sciences (PIEAS)Department of Computer Science, Tissue Image Analytics (TIA) Lab, University of WarwickPattern Recognition Lab, DCIS, Pakistan Institute of Engineering and Applied Sciences (PIEAS)Department of Physics, Islamia College PeshawarAbstract The mitotic activity index is a key prognostic measure in tumour grading. Microscopy based detection of mitotic nuclei is a significant overhead and necessitates automation. This work proposes deep CNN based multi-phase mitosis detection framework “MP-MitDet” for mitotic nuclei identification in breast cancer histopathological images. The workflow constitutes: (1) label-refiner, (2) tissue-level mitotic region selection, (3) blob analysis, and (4) cell-level refinement. We developed an automatic label-refiner to represent weak labels with semi-sematic information for training of deep CNNs. A deep instance-based detection and segmentation model is used to explore probable mitotic regions on tissue patches. More probable regions are screened based on blob area and then analysed at cell-level by developing a custom CNN classifier “MitosRes-CNN” to filter false mitoses. The performance of the proposed “MitosRes-CNN” is compared with the state-of-the-art CNNs that are adapted to cell-level discrimination through cross-domain transfer learning and by adding task-specific layers. The performance of the proposed framework shows good discrimination ability in terms of F-score (0.75), recall (0.76), precision (0.71) and area under the precision-recall curve (0.78) on challenging TUPAC16 dataset. Promising results suggest good generalization of the proposed framework that can learn characteristic features from heterogenous mitotic nuclei.https://doi.org/10.1038/s41598-021-85652-1
collection DOAJ
language English
format Article
sources DOAJ
author Anabia Sohail
Asifullah Khan
Noorul Wahab
Aneela Zameer
Saranjam Khan
spellingShingle Anabia Sohail
Asifullah Khan
Noorul Wahab
Aneela Zameer
Saranjam Khan
A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
Scientific Reports
author_facet Anabia Sohail
Asifullah Khan
Noorul Wahab
Aneela Zameer
Saranjam Khan
author_sort Anabia Sohail
title A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
title_short A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
title_full A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
title_fullStr A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
title_full_unstemmed A multi-phase deep CNN based mitosis detection framework for breast cancer histopathological images
title_sort multi-phase deep cnn based mitosis detection framework for breast cancer histopathological images
publisher Nature Publishing Group
series Scientific Reports
issn 2045-2322
publishDate 2021-03-01
description Abstract The mitotic activity index is a key prognostic measure in tumour grading. Microscopy based detection of mitotic nuclei is a significant overhead and necessitates automation. This work proposes deep CNN based multi-phase mitosis detection framework “MP-MitDet” for mitotic nuclei identification in breast cancer histopathological images. The workflow constitutes: (1) label-refiner, (2) tissue-level mitotic region selection, (3) blob analysis, and (4) cell-level refinement. We developed an automatic label-refiner to represent weak labels with semi-sematic information for training of deep CNNs. A deep instance-based detection and segmentation model is used to explore probable mitotic regions on tissue patches. More probable regions are screened based on blob area and then analysed at cell-level by developing a custom CNN classifier “MitosRes-CNN” to filter false mitoses. The performance of the proposed “MitosRes-CNN” is compared with the state-of-the-art CNNs that are adapted to cell-level discrimination through cross-domain transfer learning and by adding task-specific layers. The performance of the proposed framework shows good discrimination ability in terms of F-score (0.75), recall (0.76), precision (0.71) and area under the precision-recall curve (0.78) on challenging TUPAC16 dataset. Promising results suggest good generalization of the proposed framework that can learn characteristic features from heterogenous mitotic nuclei.
url https://doi.org/10.1038/s41598-021-85652-1
work_keys_str_mv AT anabiasohail amultiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT asifullahkhan amultiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT noorulwahab amultiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT aneelazameer amultiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT saranjamkhan amultiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT anabiasohail multiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT asifullahkhan multiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT noorulwahab multiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT aneelazameer multiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
AT saranjamkhan multiphasedeepcnnbasedmitosisdetectionframeworkforbreastcancerhistopathologicalimages
_version_ 1724210255987998720