EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automated recognition and segmentation of lung cancer cytological images based on deep learning

Qingyang Wang iD, Yazhi Luo, Ying Zhao, Shuhao Wang, Yiru Niu, Jinxi Di, Jia Guo, Guorong Lan, Lei Yang, Yu Shan Mao, Yuan Tu, Dingrong Zhong und 1 weitere

PLOS ONE · 2025

DOI: 10.1371/journal.pone.0317996 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, 2025. Verfasst von Qingyang Wang und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology, Chengdu Second People's Hospital, Sichuan, China. Hauptthemen laut MEDLINE: Lung Neoplasms, Deep Learning, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: China-Japan Friendship Hospital Youth Science and Technology Excellence Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Compared with histological examination of lung cancer, cytology is less invasive and provides better preservation of complete morphology and detail. However, traditional cytological diagnosis requires an experienced pathologist to evaluate all sections individually under a microscope, which is a time-consuming process with low interobserver consistency. With the development of deep neural networks, the You Only Look Once (YOLO) object-detection model has been recognized for its impressive speed and accuracy. Thus, in this study, we developed a model for intraoperative cytological segmentation of pulmonary lesions based on the YOLOv8 algorithm, which labels each instance by segmenting the image at the pixel level. The model achieved a mean pixel accuracy and mean intersection over union of 0.80 and 0.70, respectively, on the test set. At the image level, the accuracy and area under the receiver operating characteristic curve values for malignant and benign (or normal) lesions were 91.0% and 0.90, respectively. In addition, the model was deemed suitable for diagnosing pleural fluid cytology and bronchoalveolar lavage fluid cytology images. The model predictions were strongly correlated with pathologist diagnoses and the gold standard, indicating the model’s ability to make clinical-level decisions during initial diagnosis. Thus, the proposed method is useful for rapidly localizing lung cancer cells based on microscopic images and outputting image interpretation results.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Lung NeoplasmsDeep LearningImage Processing, Computer-AssistedHumansAlgorithmsCytodiagnosisROC CurveBronchoalveolar Lavage FluidNeural Networks, Computer

Autor:innen und Institutionen

  • Department of Pathology, Chengdu Second People's Hospital, Sichuan, China.
  • Department of Pathology, China-Japan Friendship Hospital, Beijing, China.
  • Technical University of Munich, Munich, Germany.
  • School of Optics and Photonics, Beijing Institute of Technology, Beijing, China.
  • Key Laboratory of Photoelectronic Imaging Technology and System, Ministry of Education, Beijing Institute of Technology, Beijing, China.
  • Thorough Lab, Thorough Future, Beijing, China.

ORCID-Profile: Qingyang Wang

Förderung

  • China-Japan Friendship Hospital Youth Science and Technology Excellence Project

Zitieren

APA

Wang, Q., Luo, Y., Zhao, Y., Wang, S., Niu, Y., Di, J., Guo, J., Lan, G., Yang, L., Mao, Y. S., Tu, Y., Zhong, D., & Zhang, P. (2025). Automated recognition and segmentation of lung cancer cytological images based on deep learning. PLOS ONE. https://doi.org/10.1371/journal.pone.0317996

Vancouver

Wang Q, Luo Y, Zhao Y, Wang S, Niu Y, Di J, et al. Automated recognition and segmentation of lung cancer cytological images based on deep learning. PLOS ONE. 2025. doi:10.1371/journal.pone.0317996

BibTeX
@article{Wang2025,
  author = {Wang, Qingyang and Luo, Yazhi and Zhao, Ying and Wang, Shuhao and Niu, Yiru and Di, Jinxi and Guo, Jia and Lan, Guorong and Yang, Lei and Mao, Yu Shan and Tu, Yuan and Zhong, Dingrong and Zhang, Pei},
  title = {Automated recognition and segmentation of lung cancer cytological images based on deep learning},
  journal = {PLOS ONE},
  year = {2025},
  doi = {10.1371/journal.pone.0317996}
}
Kostenlos starten →