High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network

Photoacoustic imaging is a powerful technique that provides high-resolution, deep tissue imaging. However, the time-intensive nature of photoacoustic microscopy (PAM) poses a significant challenge, especially when high-resolution images are required for real-time applications. In this study, we prop...

Full description

Saved in:
Bibliographic Details
Main Authors: Yameng Zhang, Hua Tian, Min Wan, Shihao Tang, Ziyun Ding, Wei Huang, Yamin Yang, Weitao Li
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Photoacoustics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2213597925000436
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850260377851396096
author Yameng Zhang
Hua Tian
Min Wan
Shihao Tang
Ziyun Ding
Wei Huang
Yamin Yang
Weitao Li
author_facet Yameng Zhang
Hua Tian
Min Wan
Shihao Tang
Ziyun Ding
Wei Huang
Yamin Yang
Weitao Li
author_sort Yameng Zhang
collection DOAJ
description Photoacoustic imaging is a powerful technique that provides high-resolution, deep tissue imaging. However, the time-intensive nature of photoacoustic microscopy (PAM) poses a significant challenge, especially when high-resolution images are required for real-time applications. In this study, we proposed an optimized Fast Residual Dense Generative Adversarial Network (FRDGAN) for high-quality PAM reconstruction. Through dataset validation on mouse ear vasculature, FRDGAN demonstrated superior performance in image quality, background noise suppression, and computational efficiency across multiple down-sampling scales (×4, ×8) compared to classical methods. Furthermore, in the in vivo experiments of mouse cerebral vasculature, FRDGAN achieves the improvement of 2.24 dB and 0.0255 in peak signal-to-noise ratio and structural similarity metrics in contrast to SRGAN, respectively. Our FRDGAN method provides a promising solution for fast, high-quality PAM microvascular imaging in biomedical research.
format Article
id doaj-art-cdf5360fdbc643e6aa563ef9530c590b
institution OA Journals
issn 2213-5979
language English
publishDate 2025-06-01
publisher Elsevier
record_format Article
series Photoacoustics
spelling doaj-art-cdf5360fdbc643e6aa563ef9530c590b2025-08-20T01:55:38ZengElsevierPhotoacoustics2213-59792025-06-014310072010.1016/j.pacs.2025.100720High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial networkYameng Zhang0Hua Tian1Min Wan2Shihao Tang3Ziyun Ding4Wei Huang5Yamin Yang6Weitao Li7School of Computer Engineering, Nanjing Institute of Technology, Nanjing, Jiangsu 211167, China; Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211106, ChinaSchool of Computer Engineering, Nanjing Institute of Technology, Nanjing, Jiangsu 211167, ChinaDepartment of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211106, ChinaDepartment of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211106, ChinaSchool of Engineering, University of Birmingham, Birmingham B15 2TT, UKSchool of Computer Engineering, Nanjing Institute of Technology, Nanjing, Jiangsu 211167, ChinaDepartment of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211106, ChinaDepartment of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu 211106, China; Corresponding author.Photoacoustic imaging is a powerful technique that provides high-resolution, deep tissue imaging. However, the time-intensive nature of photoacoustic microscopy (PAM) poses a significant challenge, especially when high-resolution images are required for real-time applications. In this study, we proposed an optimized Fast Residual Dense Generative Adversarial Network (FRDGAN) for high-quality PAM reconstruction. Through dataset validation on mouse ear vasculature, FRDGAN demonstrated superior performance in image quality, background noise suppression, and computational efficiency across multiple down-sampling scales (×4, ×8) compared to classical methods. Furthermore, in the in vivo experiments of mouse cerebral vasculature, FRDGAN achieves the improvement of 2.24 dB and 0.0255 in peak signal-to-noise ratio and structural similarity metrics in contrast to SRGAN, respectively. Our FRDGAN method provides a promising solution for fast, high-quality PAM microvascular imaging in biomedical research.http://www.sciencedirect.com/science/article/pii/S2213597925000436Photoacoustic microscopyGenerative adversarial networkVascular imagingSuper-resolution imagingResidual dense module
spellingShingle Yameng Zhang
Hua Tian
Min Wan
Shihao Tang
Ziyun Ding
Wei Huang
Yamin Yang
Weitao Li
High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
Photoacoustics
Photoacoustic microscopy
Generative adversarial network
Vascular imaging
Super-resolution imaging
Residual dense module
title High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
title_full High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
title_fullStr High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
title_full_unstemmed High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
title_short High resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
title_sort high resolution photoacoustic vascular image reconstruction through the fast residual dense generative adversarial network
topic Photoacoustic microscopy
Generative adversarial network
Vascular imaging
Super-resolution imaging
Residual dense module
url http://www.sciencedirect.com/science/article/pii/S2213597925000436
work_keys_str_mv AT yamengzhang highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT huatian highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT minwan highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT shihaotang highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT ziyunding highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT weihuang highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT yaminyang highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork
AT weitaoli highresolutionphotoacousticvascularimagereconstructionthroughthefastresidualdensegenerativeadversarialnetwork