Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy

Imaging speed is critical for photoacoustic microscopy as it affects the capability to capture dynamic biological processes and support real-time clinical applications. Conventional approaches for increasing imaging speed typically involve high-repetition-rate lasers, which pose a risk of thermal da...

Full description

Saved in:
Bibliographic Details
Main Authors: Yihan Pi, Jijing Chen, Kaixuan Ding, Tongyan Zhang, Hao Zhang, Bingxue Zhang, Junhao Guo, Zhen Tian, Jiao Li
Format: Article
Language:English
Published: Elsevier 2025-04-01
Series:Photoacoustics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2213597925000163
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850046230896312320
author Yihan Pi
Jijing Chen
Kaixuan Ding
Tongyan Zhang
Hao Zhang
Bingxue Zhang
Junhao Guo
Zhen Tian
Jiao Li
author_facet Yihan Pi
Jijing Chen
Kaixuan Ding
Tongyan Zhang
Hao Zhang
Bingxue Zhang
Junhao Guo
Zhen Tian
Jiao Li
author_sort Yihan Pi
collection DOAJ
description Imaging speed is critical for photoacoustic microscopy as it affects the capability to capture dynamic biological processes and support real-time clinical applications. Conventional approaches for increasing imaging speed typically involve high-repetition-rate lasers, which pose a risk of thermal damage to samples. Here, we propose a deep-learning-driven optical-scanning undersampling method for photoacoustic remote sensing (PARS) microscopy, accelerating imaging acquisition while maintaining a constant laser repetition rate and reducing laser dosage. We develop a hybrid Transformer-Convolutional Neural Network, HTC-GAN, to address the challenges of both nonuniform sampling and motion misalignment inherent in optical-scanning undersampling. A mouse ear vasculature image dataset is created through our customized galvanometer-scanned PARS system to train and validate HTC-GAN. The network successfully restores high-quality images from 1/2-undersampled and 1/4-undersampled data, closely approximating the ground truth images. A series of performance experiments demonstrate that HTC-GAN surpasses the basic misalignment compensation algorithm, and standalone CNN or Transformer networks in terms of perceptual quality and quantitative metrics. Moreover, three-dimensional imaging results validate the robustness and versatility of the proposed optical-scanning undersampling imaging method across multiscale scanning modes. Our method achieves a fourfold improvement in PARS imaging speed without hardware upgrades, offering an available solution for enhancing imaging speed in other optical-scanning microscopic systems.
format Article
id doaj-art-af7ca5620d2f4f0e8ef65cfed1181690
institution DOAJ
issn 2213-5979
language English
publishDate 2025-04-01
publisher Elsevier
record_format Article
series Photoacoustics
spelling doaj-art-af7ca5620d2f4f0e8ef65cfed11816902025-08-20T02:54:29ZengElsevierPhotoacoustics2213-59792025-04-014210069710.1016/j.pacs.2025.100697Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopyYihan Pi0Jijing Chen1Kaixuan Ding2Tongyan Zhang3Hao Zhang4Bingxue Zhang5Junhao Guo6Zhen Tian7Jiao Li8College of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; Center for Terahertz Waves, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; Center for Terahertz Waves, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, ChinaInfectious Diseases Department, Second Affiliated Hospital of Tianjin University of Traditional Chinese Medicine, Tianjin, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; Center for Terahertz Waves, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; Center for Terahertz Waves, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; Center for Terahertz Waves, Tianjin University, Tianjin 300072, China; State Key Laboratory of Precision Measurement Technology and Instruments, Tianjin University, Tianjin 300072, China; Georgia Tech Shenzhen Institute (GTSI), Tianjin University, Shenzhen 518067, China; Corresponding authors at: College of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, ChinaCollege of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, China; State Key Laboratory of Precision Measurement Technology and Instruments, Tianjin University, Tianjin 300072, China; Georgia Tech Shenzhen Institute (GTSI), Tianjin University, Shenzhen 518067, China; Corresponding authors at: College of Precision Instrument and Optoelectronics Engineering, Tianjin University, Tianjin 300072, ChinaImaging speed is critical for photoacoustic microscopy as it affects the capability to capture dynamic biological processes and support real-time clinical applications. Conventional approaches for increasing imaging speed typically involve high-repetition-rate lasers, which pose a risk of thermal damage to samples. Here, we propose a deep-learning-driven optical-scanning undersampling method for photoacoustic remote sensing (PARS) microscopy, accelerating imaging acquisition while maintaining a constant laser repetition rate and reducing laser dosage. We develop a hybrid Transformer-Convolutional Neural Network, HTC-GAN, to address the challenges of both nonuniform sampling and motion misalignment inherent in optical-scanning undersampling. A mouse ear vasculature image dataset is created through our customized galvanometer-scanned PARS system to train and validate HTC-GAN. The network successfully restores high-quality images from 1/2-undersampled and 1/4-undersampled data, closely approximating the ground truth images. A series of performance experiments demonstrate that HTC-GAN surpasses the basic misalignment compensation algorithm, and standalone CNN or Transformer networks in terms of perceptual quality and quantitative metrics. Moreover, three-dimensional imaging results validate the robustness and versatility of the proposed optical-scanning undersampling imaging method across multiscale scanning modes. Our method achieves a fourfold improvement in PARS imaging speed without hardware upgrades, offering an available solution for enhancing imaging speed in other optical-scanning microscopic systems.http://www.sciencedirect.com/science/article/pii/S2213597925000163PhotoacousticDeep learningOptical scanningUndersampling
spellingShingle Yihan Pi
Jijing Chen
Kaixuan Ding
Tongyan Zhang
Hao Zhang
Bingxue Zhang
Junhao Guo
Zhen Tian
Jiao Li
Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
Photoacoustics
Photoacoustic
Deep learning
Optical scanning
Undersampling
title Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
title_full Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
title_fullStr Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
title_full_unstemmed Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
title_short Hybrid transformer-CNN network-driven optical-scanning undersampling for photoacoustic remote sensing microscopy
title_sort hybrid transformer cnn network driven optical scanning undersampling for photoacoustic remote sensing microscopy
topic Photoacoustic
Deep learning
Optical scanning
Undersampling
url http://www.sciencedirect.com/science/article/pii/S2213597925000163
work_keys_str_mv AT yihanpi hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT jijingchen hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT kaixuanding hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT tongyanzhang hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT haozhang hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT bingxuezhang hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT junhaoguo hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT zhentian hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy
AT jiaoli hybridtransformercnnnetworkdrivenopticalscanningundersamplingforphotoacousticremotesensingmicroscopy