StainNet: A Fast and Robust Stain Normalization Network

Stain normalization often refers to transferring the color distribution to the target image and has been widely used in biomedical image analysis. The conventional stain normalization usually achieves through a pixel-by-pixel color mapping model, which depends on one reference image, and it is hard...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hongtao Kang, Die Luo, Weihua Feng, Shaoqun Zeng, Tingwei Quan, Junbo Hu, Xiuli Liu
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	stain normalization cytopathology histopathology convolutional neural network (CNN) generative adversarial network (GANs) Medicine (General) R5-920
Acceso en línea:	https://doaj.org/article/dac4b76d32094411b2305ff714fa655c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:dac4b76d32094411b2305ff714fa655c
record_format	dspace
spelling	oai:doaj.org-article:dac4b76d32094411b2305ff714fa655c2021-11-05T09:14:01ZStainNet: A Fast and Robust Stain Normalization Network2296-858X10.3389/fmed.2021.746307https://doaj.org/article/dac4b76d32094411b2305ff714fa655c2021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fmed.2021.746307/fullhttps://doaj.org/toc/2296-858XStain normalization often refers to transferring the color distribution to the target image and has been widely used in biomedical image analysis. The conventional stain normalization usually achieves through a pixel-by-pixel color mapping model, which depends on one reference image, and it is hard to achieve accurately the style transformation between image datasets. In principle, this difficulty can be well-solved by deep learning-based methods, whereas, its complicated structure results in low computational efficiency and artifacts in the style transformation, which has restricted the practical application. Here, we use distillation learning to reduce the complexity of deep learning methods and a fast and robust network called StainNet to learn the color mapping between the source image and the target image. StainNet can learn the color mapping relationship from a whole dataset and adjust the color value in a pixel-to-pixel manner. The pixel-to-pixel manner restricts the network size and avoids artifacts in the style transformation. The results on the cytopathology and histopathology datasets show that StainNet can achieve comparable performance to the deep learning-based methods. Computation results demonstrate StainNet is more than 40 times faster than StainGAN and can normalize a 100,000 × 100,000 whole slide image in 40 s.Hongtao KangHongtao KangDie LuoDie LuoWeihua FengWeihua FengShaoqun ZengShaoqun ZengTingwei QuanTingwei QuanJunbo HuXiuli LiuXiuli LiuFrontiers Media S.A.articlestain normalizationcytopathologyhistopathologyconvolutional neural network (CNN)generative adversarial network (GANs)Medicine (General)R5-920ENFrontiers in Medicine, Vol 8 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	stain normalization cytopathology histopathology convolutional neural network (CNN) generative adversarial network (GANs) Medicine (General) R5-920
spellingShingle	stain normalization cytopathology histopathology convolutional neural network (CNN) generative adversarial network (GANs) Medicine (General) R5-920 Hongtao Kang Hongtao Kang Die Luo Die Luo Weihua Feng Weihua Feng Shaoqun Zeng Shaoqun Zeng Tingwei Quan Tingwei Quan Junbo Hu Xiuli Liu Xiuli Liu StainNet: A Fast and Robust Stain Normalization Network
description	Stain normalization often refers to transferring the color distribution to the target image and has been widely used in biomedical image analysis. The conventional stain normalization usually achieves through a pixel-by-pixel color mapping model, which depends on one reference image, and it is hard to achieve accurately the style transformation between image datasets. In principle, this difficulty can be well-solved by deep learning-based methods, whereas, its complicated structure results in low computational efficiency and artifacts in the style transformation, which has restricted the practical application. Here, we use distillation learning to reduce the complexity of deep learning methods and a fast and robust network called StainNet to learn the color mapping between the source image and the target image. StainNet can learn the color mapping relationship from a whole dataset and adjust the color value in a pixel-to-pixel manner. The pixel-to-pixel manner restricts the network size and avoids artifacts in the style transformation. The results on the cytopathology and histopathology datasets show that StainNet can achieve comparable performance to the deep learning-based methods. Computation results demonstrate StainNet is more than 40 times faster than StainGAN and can normalize a 100,000 × 100,000 whole slide image in 40 s.
format	article
author	Hongtao Kang Hongtao Kang Die Luo Die Luo Weihua Feng Weihua Feng Shaoqun Zeng Shaoqun Zeng Tingwei Quan Tingwei Quan Junbo Hu Xiuli Liu Xiuli Liu
author_facet	Hongtao Kang Hongtao Kang Die Luo Die Luo Weihua Feng Weihua Feng Shaoqun Zeng Shaoqun Zeng Tingwei Quan Tingwei Quan Junbo Hu Xiuli Liu Xiuli Liu
author_sort	Hongtao Kang
title	StainNet: A Fast and Robust Stain Normalization Network
title_short	StainNet: A Fast and Robust Stain Normalization Network
title_full	StainNet: A Fast and Robust Stain Normalization Network
title_fullStr	StainNet: A Fast and Robust Stain Normalization Network
title_full_unstemmed	StainNet: A Fast and Robust Stain Normalization Network
title_sort	stainnet: a fast and robust stain normalization network
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/dac4b76d32094411b2305ff714fa655c
work_keys_str_mv	AT hongtaokang stainnetafastandrobuststainnormalizationnetwork AT hongtaokang stainnetafastandrobuststainnormalizationnetwork AT dieluo stainnetafastandrobuststainnormalizationnetwork AT dieluo stainnetafastandrobuststainnormalizationnetwork AT weihuafeng stainnetafastandrobuststainnormalizationnetwork AT weihuafeng stainnetafastandrobuststainnormalizationnetwork AT shaoqunzeng stainnetafastandrobuststainnormalizationnetwork AT shaoqunzeng stainnetafastandrobuststainnormalizationnetwork AT tingweiquan stainnetafastandrobuststainnormalizationnetwork AT tingweiquan stainnetafastandrobuststainnormalizationnetwork AT junbohu stainnetafastandrobuststainnormalizationnetwork AT xiuliliu stainnetafastandrobuststainnormalizationnetwork AT xiuliliu stainnetafastandrobuststainnormalizationnetwork
_version_	1718444440084807680

StainNet: A Fast and Robust Stain Normalization Network

Ejemplares similares