PSIG-GAN: a polarization structure information guided generative adversarial network for stain transformation

Jiahao Fan; Honghui He; Hui Ma

doi:10.1117/12.3048494

5 December 2024 PSIG-GAN: a polarization structure information guided generative adversarial network for stain transformation

Jiahao Fan, Honghui He, Hui Ma

Author Affiliations +

Proceedings Volume 13418, Fifteenth International Conference on Information Optics and Photonics (CIOP 2024); 134181R (2024) https://doi.org/10.1117/12.3048494
Event: 15th International Conference on Information Optics and Photonics (CIOP2024), 2024, Xi’an, China

Abstract

Histopathological examination is one of the important methods in clinical pathology diagnosis, traditionally involving the observation and analysis of stained tissue slices under a microscope. However, the staining process for tissue slices is typically time-consuming and resource-intensive. Therefore, with the advancement of deep learning, virtual staining technique has been proposed to address this issue. This technique aims to achieve rapid pathological diagnosis by bypassing the conventional chemical staining process. The implementation of virtual staining often relies on generative adversarial network trained on non-paired data. However, due to the lack of effective constraint information, the network often fails to achieve accurate virtual staining. Thus, our study proposes a virtual staining paradigm guided by Mueller matrix polarization information to address this issue. By introducing additional polarization information, the model can effectively learn the mapping relationship between different stained images to achieve accurate staining effects. In this study, we use CycleGAN as an example to analyze the shortcomings of generative adversarial learning based on non-paired data. Additionally, by studying the transformation between H&E-stained images and Masson trichrome (MT) stained images, we propose a polarization structure information guided generative adversarial network (PSIG-GAN). Experimental results demonstrate that by incorporating polarization information into the generative network, the model can effectively produce highly accurate virtual stained images.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Jiahao Fan, Honghui He, and Hui Ma "PSIG-GAN: a polarization structure information guided generative adversarial network for stain transformation", Proc. SPIE 13418, Fifteenth International Conference on Information Optics and Photonics (CIOP 2024), 134181R (5 December 2024); https://doi.org/10.1117/12.3048494

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
12 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Tissues

Polarization

Mueller matrices

Image processing

Polarization imaging

Adversarial training

Biological samples

Keywords/Phrases

Search In:

Publication Years