A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores

Abstract We examined whether a tool for determining Johnsen scores automatically using artificial intelligence (AI) could be used in place of traditional Johnsen scoring to support pathologists’ evaluations. Average precision, precision, and recall were assessed by the Google Cloud AutoML Vision pla...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yurika Ito, Mami Unagami, Fumito Yamabe, Yozo Mitsui, Koichi Nakajima, Koichi Nagao, Hideyuki Kobayashi
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/56570af72ab6480f9c1c1f41d77d342e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:56570af72ab6480f9c1c1f41d77d342e
record_format	dspace
spelling	oai:doaj.org-article:56570af72ab6480f9c1c1f41d77d342e2021-12-02T15:36:13ZA method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores10.1038/s41598-021-89369-z2045-2322https://doaj.org/article/56570af72ab6480f9c1c1f41d77d342e2021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-89369-zhttps://doaj.org/toc/2045-2322Abstract We examined whether a tool for determining Johnsen scores automatically using artificial intelligence (AI) could be used in place of traditional Johnsen scoring to support pathologists’ evaluations. Average precision, precision, and recall were assessed by the Google Cloud AutoML Vision platform. We obtained testicular tissues for 275 patients and were able to use haematoxylin and eosin (H&E)-stained glass microscope slides from 264 patients. In addition, we cut out of parts of the histopathology images (5.0 × 5.0 cm) for expansion of Johnsen’s characteristic areas with seminiferous tubules. We defined four labels: Johnsen score 1–3, 4–5, 6–7, and 8–10 to distinguish Johnsen scores in clinical practice. All images were uploaded to the Google Cloud AutoML Vision platform. We obtained a dataset of 7155 images at magnification 400× and a dataset of 9822 expansion images for the 5.0 × 5.0 cm cutouts. For the 400× magnification image dataset, the average precision (positive predictive value) of the algorithm was 82.6%, precision was 80.31%, and recall was 60.96%. For the expansion image dataset (5.0 × 5.0 cm), the average precision was 99.5%, precision was 96.29%, and recall was 96.23%. This is the first report of an AI-based algorithm for predicting Johnsen scores.Yurika ItoMami UnagamiFumito YamabeYozo MitsuiKoichi NakajimaKoichi NagaoHideyuki KobayashiNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Yurika Ito Mami Unagami Fumito Yamabe Yozo Mitsui Koichi Nakajima Koichi Nagao Hideyuki Kobayashi A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
description	Abstract We examined whether a tool for determining Johnsen scores automatically using artificial intelligence (AI) could be used in place of traditional Johnsen scoring to support pathologists’ evaluations. Average precision, precision, and recall were assessed by the Google Cloud AutoML Vision platform. We obtained testicular tissues for 275 patients and were able to use haematoxylin and eosin (H&E)-stained glass microscope slides from 264 patients. In addition, we cut out of parts of the histopathology images (5.0 × 5.0 cm) for expansion of Johnsen’s characteristic areas with seminiferous tubules. We defined four labels: Johnsen score 1–3, 4–5, 6–7, and 8–10 to distinguish Johnsen scores in clinical practice. All images were uploaded to the Google Cloud AutoML Vision platform. We obtained a dataset of 7155 images at magnification 400× and a dataset of 9822 expansion images for the 5.0 × 5.0 cm cutouts. For the 400× magnification image dataset, the average precision (positive predictive value) of the algorithm was 82.6%, precision was 80.31%, and recall was 60.96%. For the expansion image dataset (5.0 × 5.0 cm), the average precision was 99.5%, precision was 96.29%, and recall was 96.23%. This is the first report of an AI-based algorithm for predicting Johnsen scores.
format	article
author	Yurika Ito Mami Unagami Fumito Yamabe Yozo Mitsui Koichi Nakajima Koichi Nagao Hideyuki Kobayashi
author_facet	Yurika Ito Mami Unagami Fumito Yamabe Yozo Mitsui Koichi Nakajima Koichi Nagao Hideyuki Kobayashi
author_sort	Yurika Ito
title	A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
title_short	A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
title_full	A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
title_fullStr	A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
title_full_unstemmed	A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores
title_sort	method for utilizing automated machine learning for histopathological classification of testis based on johnsen scores
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/56570af72ab6480f9c1c1f41d77d342e
work_keys_str_mv	AT yurikaito amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT mamiunagami amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT fumitoyamabe amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT yozomitsui amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT koichinakajima amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT koichinagao amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT hideyukikobayashi amethodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT yurikaito methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT mamiunagami methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT fumitoyamabe methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT yozomitsui methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT koichinakajima methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT koichinagao methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores AT hideyukikobayashi methodforutilizingautomatedmachinelearningforhistopathologicalclassificationoftestisbasedonjohnsenscores
_version_	1718386353225334784

A method for utilizing automated machine learning for histopathological classification of testis based on Johnsen scores

Ejemplares similares