Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector

Despite the prognostic importance of mitotic count as one of the components of the Bloom – Richardson grade [3], several studies ([2, 9, 10]) have found that pathologists’ agreement on the mitotic grade is fairly modest. Collecting a set of more than 4,200 candidate mitotic figures, we evaluate path...

Full description

Saved in:
Bibliographic Details
Main Authors: Christopher Malon, Elena Brachtel, Eric Cosatto, Hans Peter Graf, Atsushi Kurata, Masahiko Kuroda, John S. Meyer, Akira Saito, Shulin Wu, Yukako Yagi
Format: Article
Language:English
Published: Wiley 2012-01-01
Series:Analytical Cellular Pathology
Online Access:http://dx.doi.org/10.3233/ACP-2011-0029
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850168569790201856
author Christopher Malon
Elena Brachtel
Eric Cosatto
Hans Peter Graf
Atsushi Kurata
Masahiko Kuroda
John S. Meyer
Akira Saito
Shulin Wu
Yukako Yagi
author_facet Christopher Malon
Elena Brachtel
Eric Cosatto
Hans Peter Graf
Atsushi Kurata
Masahiko Kuroda
John S. Meyer
Akira Saito
Shulin Wu
Yukako Yagi
author_sort Christopher Malon
collection DOAJ
description Despite the prognostic importance of mitotic count as one of the components of the Bloom – Richardson grade [3], several studies ([2, 9, 10]) have found that pathologists’ agreement on the mitotic grade is fairly modest. Collecting a set of more than 4,200 candidate mitotic figures, we evaluate pathologists' agreement on individual figures, and train a computerized system for mitosis detection, comparing its performance to the classifications of three pathologists. The system’s and the pathologists’ classifications are based on evaluation of digital micrographs of hematoxylin and eosin stained breast tissue. On figures where the majority of pathologists agree on a classification, we compare the performance of the trained system to that of the individual pathologists. We find that the level of agreement of the pathologists ranges from slight to moderate, with strong biases, and that the system performs competitively in rating the ground truth set. This study is a step towards automatic mitosis count to accelerate a pathologist's work and improve reproducibility.
format Article
id doaj-art-a779ca306ffb4affb30d9a1834e11fa9
institution OA Journals
issn 2210-7177
2210-7185
language English
publishDate 2012-01-01
publisher Wiley
record_format Article
series Analytical Cellular Pathology
spelling doaj-art-a779ca306ffb4affb30d9a1834e11fa92025-08-20T02:20:56ZengWileyAnalytical Cellular Pathology2210-71772210-71852012-01-013529710010.3233/ACP-2011-0029Mitotic Figure Recognition: Agreement among Pathologists and Computerized DetectorChristopher Malon0Elena Brachtel1Eric Cosatto2Hans Peter Graf3Atsushi Kurata4Masahiko Kuroda5John S. Meyer6Akira Saito7Shulin Wu8Yukako Yagi9Department of Machine Learning, NEC Laboratories America, NJ, USADepartment of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USADepartment of Machine Learning, NEC Laboratories America, NJ, USADepartment of Machine Learning, NEC Laboratories America, NJ, USADepartment of Molecular Pathology, Tokyo Medical University, Tokyo, JapanDepartment of Molecular Pathology, Tokyo Medical University, Tokyo, JapanDepartment of Pathology, St. Luke's Hospital (St. Louis), Chesterfield, MO, USAInnovative Service Solutions Division, NEC Corporation, Tokyo, JapanDepartment of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USADepartment of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USADespite the prognostic importance of mitotic count as one of the components of the Bloom – Richardson grade [3], several studies ([2, 9, 10]) have found that pathologists’ agreement on the mitotic grade is fairly modest. Collecting a set of more than 4,200 candidate mitotic figures, we evaluate pathologists' agreement on individual figures, and train a computerized system for mitosis detection, comparing its performance to the classifications of three pathologists. The system’s and the pathologists’ classifications are based on evaluation of digital micrographs of hematoxylin and eosin stained breast tissue. On figures where the majority of pathologists agree on a classification, we compare the performance of the trained system to that of the individual pathologists. We find that the level of agreement of the pathologists ranges from slight to moderate, with strong biases, and that the system performs competitively in rating the ground truth set. This study is a step towards automatic mitosis count to accelerate a pathologist's work and improve reproducibility.http://dx.doi.org/10.3233/ACP-2011-0029
spellingShingle Christopher Malon
Elena Brachtel
Eric Cosatto
Hans Peter Graf
Atsushi Kurata
Masahiko Kuroda
John S. Meyer
Akira Saito
Shulin Wu
Yukako Yagi
Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
Analytical Cellular Pathology
title Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
title_full Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
title_fullStr Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
title_full_unstemmed Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
title_short Mitotic Figure Recognition: Agreement among Pathologists and Computerized Detector
title_sort mitotic figure recognition agreement among pathologists and computerized detector
url http://dx.doi.org/10.3233/ACP-2011-0029
work_keys_str_mv AT christophermalon mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT elenabrachtel mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT ericcosatto mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT hanspetergraf mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT atsushikurata mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT masahikokuroda mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT johnsmeyer mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT akirasaito mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT shulinwu mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector
AT yukakoyagi mitoticfigurerecognitionagreementamongpathologistsandcomputerizeddetector