{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T11:36:53Z","timestamp":1773920213754,"version":"3.50.1"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T00:00:00Z","timestamp":1773878400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Bioeng. Biotechnol."],"abstract":"<jats:p>\n                    Real-time\n                    <jats:italic>in situ<\/jats:italic>\n                    biopsy marks a paradigm shift in clinical oncology by enabling immediate intraprocedural pathological diagnosis during endoscopy. Photoacoustic endoscopy (PAE) is a pivotal technology in this field that uniquely visualizes tumor microvasculature and hypoxia through the synergistic fusion of optical contrast and ultrasonic resolution. However, PAE\u2019s inherent resolution-speed tradeoff in raster scanning induces severe motion artifacts from physiological activities (e.g., peristalsis and respiration), critically compromising diagnostic reliability. Although deep-learning-based super-resolution (SR) techniques show promise for photoacoustic microscopy, their clinical translation to PAE is hindered by excessive computational demands and insufficient real-time performance. To overcome this limitation, we propose Light-RepViTSR, an ultra-lightweight SR reconstruction network based on the RepViT architecture and specifically optimized for real-time PAE. Our approach integrates the representational capacity of RepViT\u2019s re-parameterizable convolutional blocks while eliminating non-essential components (e.g., squeeze-and-excitation layers) to maximize computational efficiency. Comprehensive evaluation on a multi-source dataset\u2014including 19 previously unseen murine cerebrovascular images and 18 self-collected plant vein images\u2014demonstrates the superiority of Light-RepViTSR. The network consistently outperforms conventional methods across scaling factors (\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m1\">\n                        <mml:mrow>\n                          <mml:mo>\u00d7<\/mml:mo>\n                          <mml:mn>2<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ,\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m2\">\n                        <mml:mrow>\n                          <mml:mo>\u00d7<\/mml:mo>\n                          <mml:mn>4<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    , and\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m3\">\n                        <mml:mrow>\n                          <mml:mo>\u00d7<\/mml:mo>\n                          <mml:mn>8<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ) to achieve significant improvements in PSNR (up to\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m4\">\n                        <mml:mrow>\n                          <mml:mo>+<\/mml:mo>\n                          <mml:mn>1.41<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    \u00a0dB at\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m5\">\n                        <mml:mrow>\n                          <mml:mo>\u00d7<\/mml:mo>\n                          <mml:mn>8<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ) and SSIM (up to\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m6\">\n                        <mml:mrow>\n                          <mml:mo>+<\/mml:mo>\n                          <mml:mn>0.047<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    at\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m7\">\n                        <mml:mrow>\n                          <mml:mo>\u00d7<\/mml:mo>\n                          <mml:mn>2<\/mml:mn>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ) while reducing model size by\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m8\">\n                        <mml:mrow>\n                          <mml:mo>&gt;<\/mml:mo>\n                          <mml:mn>99<\/mml:mn>\n                          <mml:mi>%<\/mml:mi>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    and inference time by\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"m9\">\n                        <mml:mrow>\n                          <mml:mo>&gt;<\/mml:mo>\n                          <mml:mn>60<\/mml:mn>\n                          <mml:mi>%<\/mml:mi>\n                        <\/mml:mrow>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    versus SRResNet. This study establishes a pathway toward practical real-time high-resolution PAE, demonstrating significant potential for enhancing\n                    <jats:italic>in situ<\/jats:italic>\n                    tumor biopsy accuracy.\n                  <\/jats:p>","DOI":"10.3389\/fbioe.2026.1762967","type":"journal-article","created":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T06:48:40Z","timestamp":1773902920000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Light-RepViTSR: ultra-lightweight super-resolution for real-time photoacoustic endoscopy in tumor biopsy"],"prefix":"10.3389","volume":"14","author":[{"given":"Guanyi","family":"Jiang","sequence":"first","affiliation":[{"name":"Department of Hematology, Peking University Shenzhen Hospital","place":["Shenzhen, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rui","family":"Yang","sequence":"additional","affiliation":[{"name":"Dongguan Key Laboratory for Cross-Scale Autonomous Interventional Surgical Robotics, School of Advanced Engineering, Great Bay University","place":["Dongguan, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhanfeng","family":"Fang","sequence":"additional","affiliation":[{"name":"Dongguan Key Laboratory for Cross-Scale Autonomous Interventional Surgical Robotics, School of Advanced Engineering, Great Bay University","place":["Dongguan, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuwei","family":"Luo","sequence":"additional","affiliation":[{"name":"Dongguan Key Laboratory for Cross-Scale Autonomous Interventional Surgical Robotics, School of Advanced Engineering, Great Bay University","place":["Dongguan, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xianghu","family":"Yu","sequence":"additional","affiliation":[{"name":"Dongguan Key Laboratory for Cross-Scale Autonomous Interventional Surgical Robotics, School of Advanced Engineering, Great Bay University","place":["Dongguan, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Li","family":"Liu","sequence":"additional","affiliation":[{"name":"Dongguan Key Laboratory for Cross-Scale Autonomous Interventional Surgical Robotics, School of Advanced Engineering, Great Bay University","place":["Dongguan, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,3,19]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1408","DOI":"10.1148\/rg.2017160175","article-title":"Clinical significance of us artifacts","volume":"37","author":"Baad","year":"2017","journal-title":"Radiographics"},{"key":"B2","doi-asserted-by":"publisher","first-page":"e621","DOI":"10.7717\/peerj-cs.621","article-title":"A comprehensive review of deep learning-based single image super-resolution","volume":"7","author":"Bashir","year":"2021","journal-title":"PeerJ Comput. Sci."},{"key":"B3","doi-asserted-by":"publisher","first-page":"3773","DOI":"10.1364\/OL.44.003773","article-title":"Simultaneous photoacoustic imaging of intravascular and tissue oxygenation","volume":"44","author":"Chen","year":"2019","journal-title":"Opt. Lett."},{"key":"B4","doi-asserted-by":"publisher","first-page":"1915","DOI":"10.1007\/s11831-023-10032-z","article-title":"Review of lightweight deep convolutional neural networks","volume":"31","author":"Chen","year":"2024","journal-title":"Archives Comput. Methods Eng."},{"key":"B5","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1007\/s10238-024-01501-1","article-title":"Tumor hypoxia unveiled: insights into microenvironment, detection tools and emerging therapies","volume":"24","author":"Ciep\u0142a","year":"2024","journal-title":"Clin. Exp. Med."},{"key":"B6","doi-asserted-by":"publisher","first-page":"1905","DOI":"10.1007\/s10462-022-10213-5","article-title":"A review of convolutional neural network architectures and their optimizations","volume":"56","author":"Cong","year":"2023","journal-title":"Artif. Intell. Rev."},{"key":"B7","doi-asserted-by":"publisher","first-page":"37","DOI":"10.3390\/s24010037","article-title":"Biomarkers in cancer detection, diagnosis, and prognosis","volume":"24","author":"Das","year":"2023","journal-title":"Sensors"},{"key":"B8","doi-asserted-by":"publisher","first-page":"911","DOI":"10.1007\/s10489-025-06780-z","article-title":"Ssrepvm-unet: a lightweight hybrid model for medical image segmentation based on channel parallelism","volume":"55","author":"Guo","year":"2025","journal-title":"Appl. Intell."},{"key":"B9","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2018.00179","article-title":"Deep back-projection networks for super-resolution","author":"Haris","year":"2018"},{"key":"B10","doi-asserted-by":"publisher","first-page":"011101","DOI":"10.1117\/1.3281673","article-title":"Photoacoustic imaging and characterization of the microvasculature","volume":"15","author":"Hu","year":"2010","journal-title":"J. Biomedical Optics"},{"key":"B11","doi-asserted-by":"publisher","first-page":"5164970","DOI":"10.1155\/2022\/5164970","article-title":"Modern diagnostic imaging technique applications and risk factors in the medical field: a review","volume":"2022","author":"Hussain","year":"2022","journal-title":"BioMed Research International"},{"key":"B12","doi-asserted-by":"publisher","first-page":"6865","DOI":"10.3390\/app10196865","article-title":"Endoscopic optical imaging technologies and devices for medical purposes: state of the art","volume":"10","author":"Kaur","year":"2020","journal-title":"Appl. Sci."},{"key":"B13","doi-asserted-by":"publisher","first-page":"214","DOI":"10.1364\/optica.478211","article-title":"Opportunities of optical and spectral technologies in intraoperative histopathology","volume":"10","author":"Krafft","year":"2023","journal-title":"Optica"},{"key":"B14","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.19","article-title":"Photo-realistic single image super-resolution using a generative adversarial network","author":"Ledig","year":"2017"},{"key":"B15","doi-asserted-by":"crossref","DOI":"10.1109\/CVPRW.2017.151","volume-title":"Enhanced deep residual networks for single image super-resolution","author":"Lim","year":"2017"},{"key":"B16","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1038\/s41571-022-00615-3","article-title":"The emerging role of photoacoustic imaging in clinical oncology","volume":"19","author":"Lin","year":"2022","journal-title":"Nat. Rev. Clin. Oncol."},{"key":"B17","doi-asserted-by":"publisher","first-page":"2053","DOI":"10.1039\/c8cs00618k","article-title":"Photothermal therapy and photoacoustic imaging via nanotheranostics in fighting cancer","volume":"48","author":"Liu","year":"2019","journal-title":"Chem. Soc. Rev."},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3657282","article-title":"Lightweight deep learning for resource-constrained environments: a survey","volume":"56","author":"Liu","year":"2024","journal-title":"ACM Comput. Surv."},{"key":"B19","doi-asserted-by":"publisher","first-page":"715","DOI":"10.1007\/s13402-021-00602-3","article-title":"Angiogenesis as a hallmark of solid tumors-clinical perspectives","volume":"44","author":"Majidpoor","year":"2021","journal-title":"Cell. Oncol."},{"key":"B20","doi-asserted-by":"publisher","first-page":"497","DOI":"10.1515\/nanoph-2016-0026","article-title":"Ultrafast optical imaging technology: principles and applications of emerging methods","volume":"5","author":"Mikami","year":"2016","journal-title":"Nanophotonics"},{"key":"B21","doi-asserted-by":"publisher","first-page":"e12776","DOI":"10.1111\/micc.12776","article-title":"Photoacoustic imaging for microcirculation","volume":"29","author":"Mirg","year":"2022","journal-title":"Microcirculation"},{"key":"B22","doi-asserted-by":"publisher","first-page":"3098","DOI":"10.1007\/s00259-022-05746-4","article-title":"Deep learning-based image reconstruction and post-processing methods in positron emission tomography for low-dose imaging and resolution enhancement","volume":"49","author":"Pain","year":"2022","journal-title":"Eur. J. Nucl. Med. Mol. Imaging"},{"key":"B23","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1038\/s44222-024-00240-y","article-title":"Clinical translation of photoacoustic imaging","volume":"3","author":"Park","year":"2025","journal-title":"Nat. Rev. Bioeng."},{"key":"B24","doi-asserted-by":"publisher","first-page":"107590","DOI":"10.1016\/j.cmpb.2023.107590","article-title":"Medical image super-resolution reconstruction algorithms based on deep learning: a survey","volume":"238","author":"Qiu","year":"2023","journal-title":"Comput. Methods Programs Biomed."},{"key":"B25","doi-asserted-by":"publisher","first-page":"79","DOI":"10.3390\/bios10070079","article-title":"Skin-integrated wearable systems and implantable biosensors: a comprehensive review","volume":"10","author":"Rodrigues","year":"2020","journal-title":"Biosensors"},{"key":"B26","doi-asserted-by":"publisher","first-page":"308","DOI":"10.4161\/cbt.6.3.3994","article-title":"Molecular genetic tools shape a roadmap towards a more accurate prognostic prediction and personalized management of cancer","volume":"6","author":"Roukos","year":"2007","journal-title":"Cancer Biology and Therapy"},{"key":"B27","doi-asserted-by":"publisher","first-page":"57","DOI":"10.14694\/EdBook_AM.2015.35.57","article-title":"Early detection of cancer: past, present, and future","volume":"35","author":"Schiffman","year":"2015","journal-title":"Am. Soc. Clin. Oncol. Educ. Book"},{"key":"B28","doi-asserted-by":"publisher","first-page":"465","DOI":"10.1007\/s13534-024-00365-4","article-title":"Super-resolution techniques for biomedical applications and challenges","volume":"14","author":"Shin","year":"2024","journal-title":"Biomed. Eng. Lett."},{"key":"B29","doi-asserted-by":"publisher","first-page":"e00286","DOI":"10.1002\/lpor.202500286","article-title":"Optical multimodal endoscopy","volume":"19","author":"Song","year":"2025","journal-title":"Laser and Photonics Rev."},{"key":"B30","article-title":"Repvit-sam: towards real-time segmenting anything","author":"Wang","year":"2023","journal-title":"arXiv Preprint arXiv:2312.05760"},{"key":"B31","first-page":"15909","article-title":"Repvit: revisiting mobile cnn from vit perspective","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"Wang","year":"2024"},{"key":"B32","doi-asserted-by":"publisher","first-page":"27574","DOI":"10.1364\/OE.528474","article-title":"Parallel diffusion models promote high detail-fidelity photoacoustic microscopy in sparse sampling","volume":"32","author":"Wu","year":"2024","journal-title":"Opt. Express"},{"key":"B33","doi-asserted-by":"publisher","first-page":"20891","DOI":"10.1007\/s10489-023-04566-9","article-title":"Deep learning in medical image super resolution: a review","volume":"53","author":"Yang","year":"2023","journal-title":"Appl. Intell."},{"key":"B34","doi-asserted-by":"publisher","first-page":"534","DOI":"10.4253\/wjge.v5.i11.534","article-title":"Recent advances in photoacoustic endoscopy","volume":"5","author":"Yoon","year":"2013","journal-title":"World Journal Gastrointestinal Endoscopy"},{"key":"B35","first-page":"1","article-title":"Mobilesr: efficient convolutional neural network for super-resolution","volume-title":"Globecom 2020 - 2020 IEEE global communications conference","author":"Zhang","year":"2020"},{"key":"B36","doi-asserted-by":"publisher","first-page":"e0330678","DOI":"10.1371\/journal.pone.0330678","article-title":"Echomamba: a new mamba model for fast and efficient hyperspectral image classification","volume":"20","author":"Zhang","year":"2025","journal-title":"PloS One"},{"key":"B37","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1016\/j.mri.2019.05.038","article-title":"Applications of a deep learning method for anti-aliasing and super-resolution in mri","volume":"64","author":"Zhao","year":"2019","journal-title":"Magn. Resonance Imaging"}],"container-title":["Frontiers in Bioengineering and Biotechnology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1762967\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T06:48:42Z","timestamp":1773902922000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1762967\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,19]]},"references-count":37,"alternative-id":["10.3389\/fbioe.2026.1762967"],"URL":"https:\/\/doi.org\/10.3389\/fbioe.2026.1762967","relation":{},"ISSN":["2296-4185"],"issn-type":[{"value":"2296-4185","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,19]]},"article-number":"1762967"}}