GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection

Haoyuan Chen; Chen Li; Ge Wang; Xiaoyan Li; Md Mamunur Rahaman; Hongzan Sun; Weiming Hu; Yixin Li; Wanli Liu; Changhao Sun; Shiliang Ai; Marcin Grzegorzek

doi:10.1016/j.patcog.2022.108827

GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection

Haoyuan Chen, Chen Li^*, Ge Wang, Xiaoyan Li, Md Mamunur Rahaman, Hongzan Sun, Weiming Hu, Yixin Li, Wanli Liu, Changhao Sun, Shiliang Ai, Marcin Grzegorzek

^*Corresponding author for this work

222 Citations (Scopus)

Abstract

In this paper, a multi-scale visual transformer model, referred as GasHis-Transformer, is proposed for Gastric Histopathological Image Detection (GHID), which enables the automatic global detection of gastric cancer images. GasHis-Transformer model consists of two key modules designed to extract global and local information using a position-encoded transformer model and a convolutional neural network with local convolution, respectively. A publicly available hematoxylin and eosin (H&E) stained gastric histopathological image dataset is used in the experiment. Furthermore, a Dropconnect based lightweight network is proposed to reduce the model size and training time of GasHis-Transformer for clinical applications with improved confidence. Moreover, a series of contrast and extended experiments verify the robustness, extensibility and stability of GasHis-Transformer. In conclusion, GasHis-Transformer demonstrates high global detection performance and shows its significant potential in GHID task.

Original language	English
Article number	108827
Journal	Pattern Recognition
Volume	130
ISSN	0031-3203
DOIs	https://doi.org/10.1016/j.patcog.2022.108827
Publication status	Published - 10.2022

Funding

This work is supported by National Natural Science Foundation of China (No. 61806047 ). We thank Miss Zixian Li and Mr. Guoxian Li for their important discussion. We also thank Mr. Jinghua Zhang for his contribution to the revision of this paper. There is no conflict of interest in this paper.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 4 Quality Education
SDG 9 Industry, Innovation, and Infrastructure
SDG 11 Sustainable Cities and Communities
SDG 12 Responsible Consumption and Production
SDG 14 Life Below Water
SDG 15 Life on Land

Research Areas and Centers

Centers: Center for Artificial Intelligence Luebeck (ZKIL)

Access to Document

10.1016/j.patcog.2022.108827

Cite this

@article{3b7cc10b28ed4d3e9cf59b73b4401c65,

title = "GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection",

abstract = "In this paper, a multi-scale visual transformer model, referred as GasHis-Transformer, is proposed for Gastric Histopathological Image Detection (GHID), which enables the automatic global detection of gastric cancer images. GasHis-Transformer model consists of two key modules designed to extract global and local information using a position-encoded transformer model and a convolutional neural network with local convolution, respectively. A publicly available hematoxylin and eosin (H\&E) stained gastric histopathological image dataset is used in the experiment. Furthermore, a Dropconnect based lightweight network is proposed to reduce the model size and training time of GasHis-Transformer for clinical applications with improved confidence. Moreover, a series of contrast and extended experiments verify the robustness, extensibility and stability of GasHis-Transformer. In conclusion, GasHis-Transformer demonstrates high global detection performance and shows its significant potential in GHID task.",

author = "Haoyuan Chen and Chen Li and Ge Wang and Xiaoyan Li and \{Mamunur Rahaman\}, Md and Hongzan Sun and Weiming Hu and Yixin Li and Wanli Liu and Changhao Sun and Shiliang Ai and Marcin Grzegorzek",

note = "Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = oct,

doi = "10.1016/j.patcog.2022.108827",

language = "English",

volume = "130",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - GasHis-Transformer

T2 - A multi-scale visual transformer approach for gastric histopathological image detection

AU - Chen, Haoyuan

AU - Li, Chen

AU - Wang, Ge

AU - Li, Xiaoyan

AU - Mamunur Rahaman, Md

AU - Sun, Hongzan

AU - Hu, Weiming

AU - Li, Yixin

AU - Liu, Wanli

AU - Sun, Changhao

AU - Ai, Shiliang

AU - Grzegorzek, Marcin

PY - 2022/10

Y1 - 2022/10

N2 - In this paper, a multi-scale visual transformer model, referred as GasHis-Transformer, is proposed for Gastric Histopathological Image Detection (GHID), which enables the automatic global detection of gastric cancer images. GasHis-Transformer model consists of two key modules designed to extract global and local information using a position-encoded transformer model and a convolutional neural network with local convolution, respectively. A publicly available hematoxylin and eosin (H&E) stained gastric histopathological image dataset is used in the experiment. Furthermore, a Dropconnect based lightweight network is proposed to reduce the model size and training time of GasHis-Transformer for clinical applications with improved confidence. Moreover, a series of contrast and extended experiments verify the robustness, extensibility and stability of GasHis-Transformer. In conclusion, GasHis-Transformer demonstrates high global detection performance and shows its significant potential in GHID task.

AB - In this paper, a multi-scale visual transformer model, referred as GasHis-Transformer, is proposed for Gastric Histopathological Image Detection (GHID), which enables the automatic global detection of gastric cancer images. GasHis-Transformer model consists of two key modules designed to extract global and local information using a position-encoded transformer model and a convolutional neural network with local convolution, respectively. A publicly available hematoxylin and eosin (H&E) stained gastric histopathological image dataset is used in the experiment. Furthermore, a Dropconnect based lightweight network is proposed to reduce the model size and training time of GasHis-Transformer for clinical applications with improved confidence. Moreover, a series of contrast and extended experiments verify the robustness, extensibility and stability of GasHis-Transformer. In conclusion, GasHis-Transformer demonstrates high global detection performance and shows its significant potential in GHID task.

UR - https://www.scopus.com/pages/publications/85131457941

UR - https://doi.org/10.1016/j.patcog.2022.108827

U2 - 10.1016/j.patcog.2022.108827

DO - 10.1016/j.patcog.2022.108827

M3 - Journal articles

AN - SCOPUS:85131457941

SN - 0031-3203

VL - 130

JO - Pattern Recognition

JF - Pattern Recognition

M1 - 108827

ER -

GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection

Abstract

Funding

UN SDGs

Research Areas and Centers

Access to Document

Other files and links

Fingerprint

Cite this