EUVIMEDEUVIMED entdecken
Open Access

Rapid cancer diagnosis using deep learning–powered label-free subcellular-resolution photoacoustic histology

Byullee Park iD, Rui Cao iD, Yilin Luo iD, Cindy Liu iD, Yushun Zeng iD, Yide Zhang iD, Qifa Zhou iD, Samuel Davis iD, Massimo D'Apuzzo iD, Lihong V Wang iD

Science Advances · 21.11.2025 · 11 (47)

DOI: 10.1126/sciadv.adz1820 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, erschienen am 21.11.2025. Verfasst von Byullee Park und 9 weiteren Autor:innen. Beteiligte Einrichtung: Caltech Optical Imaging Laboratory, Andrew and Peggy Cherng Department of Medical Engineering, California Institute of Technology, Pasadena, CA, USA. Hauptthemen laut MEDLINE: Deep Learning, Photoacoustic Techniques, Neoplasms. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, NIBIB NIH HHS, NHLBI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional hematoxylin and eosin staining in formalin-fixed paraffin-embedded sections, while essential for diagnostic pathology, is time-consuming, labor intensive, and prone to artifacts that can obscure critical histological details. Label-free ultraviolet photoacoustic microscopy (UV-PAM) has emerged as a promising alternative, offering fast histology-like images without the need for traditional staining and excessive tissue preparation. However, current UV-PAM systems face challenges in achieving the high spatial resolution required for detailed histological analysis and diagnosis. To address this, we developed a subcellular-resolution UV-PAM (SRUV-PAM) system with a 240-nanometer resolution, enabled by the integration of a high numerical aperture (NA) objective lens (NA = 0.64) and the precise piezo actuators for fine scanning control. This configuration allows visualization of detailed nuclear structures. In addition, we demonstrated virtual staining of SRUV-PAM images via cycle-consistent generative adversarial networks and diagnosis of malignant and benign tumors in liver tissues via densely connected convolutional networks DenseNet-121, achieving an area under the receiver operating characteristic curve of 0.902.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningPhotoacoustic TechniquesNeoplasmsHumansROC CurveMicroscopyImage Processing, Computer-Assisted

Autor:innen und Institutionen

  • Caltech Optical Imaging Laboratory, Andrew and Peggy Cherng Department of Medical Engineering, California Institute of Technology, Pasadena, CA, USA.
  • Alfred E. Mann Department of Biomedical Engineering and Ophthalmology, Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA.
  • Department of Pathology, City of Hope Beckman Research Institute and Medical Center, Duarte, CA, USA.

ORCID-Profile: Byullee Park, Rui Cao, Yilin Luo, Cindy Liu, Yushun Zeng, Yide Zhang, Qifa Zhou, Samuel Davis, Massimo D'Apuzzo, Lihong V Wang

Förderung

  • NIGMS NIH HHS
  • NIBIB NIH HHS
  • NHLBI NIH HHS
  • NCI NIH HHS
  • National Institutes of Health
  • Ministry of Education
  • National Research Foundation of Korea

Zitieren

APA

Park, B., Cao, R., Luo, Y., Liu, C., Zeng, Y., Zhang, Y., Zhou, Q., Davis, S., D'Apuzzo, M., & Wang, L. V. (2025). Rapid cancer diagnosis using deep learning–powered label-free subcellular-resolution photoacoustic histology. Science Advances, 11(47). https://doi.org/10.1126/sciadv.adz1820

Vancouver

Park B, Cao R, Luo Y, Liu C, Zeng Y, Zhang Y, et al. Rapid cancer diagnosis using deep learning–powered label-free subcellular-resolution photoacoustic histology. Science Advances. 2025;11(47). doi:10.1126/sciadv.adz1820

BibTeX
@article{Park2025,
  author = {Park, Byullee and Cao, Rui and Luo, Yilin and Liu, Cindy and Zeng, Yushun and Zhang, Yide and Zhou, Qifa and Davis, Samuel and D'Apuzzo, Massimo and Wang, Lihong V},
  title = {Rapid cancer diagnosis using deep learning–powered label-free subcellular-resolution photoacoustic histology},
  journal = {Science Advances},
  year = {2025},
  volume = {11},
  number = {47},
  doi = {10.1126/sciadv.adz1820}
}
Kostenlos starten →