EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Deep learning neural network of adenocarcinoma detection in effusion cytology

Katsuhide Ikeda iD, Nanako Sakabe, Kenta Fukuda, Shouichi Sato, Toshiaki Hara, Harumi Kobayashi, Masato Nakaguro, Kennosuke Karube, Kohzo Nagata

American Journal of Clinical Pathology · 2025

DOI: 10.1093/ajcp/aqaf067 ↗

Kurz gefasst

Fachartikel in der Zeitschrift American Journal of Clinical Pathology, 2025. Verfasst von Katsuhide Ikeda und 8 weiteren Autor:innen. Beteiligte Einrichtung: Pathophysiology Sciences, Department of Integrated Health Sciences, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan. Hauptthemen laut MEDLINE: Deep Learning, Adenocarcinoma, Cytodiagnosis, Neural Networks, Computer. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Japan Society for the Promotion of Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Cytologic examination, which confirms the presence or absence of malignant cells, detects malignant cells from various organs, with adenocarcinoma as the most common histologic type. We developed a deep learning model to detect malignant cells in images obtained following effusion cytology.

Methods

The deep learning model was created using the YOLOv8 object detection algorithm (Roboflow, Inc) and 275 cases of adenocarcinoma comprising 12 182 images and 29 245 labels as well as 188 cases negative for malignancy comprising 1980 images.

Results

The adenocarcinoma test dataset exhibited Precision, Recall, F1, and mean average Precision scores of 0.909, 0.911, 0.910, and 0.955, respectively. The number of adenocarcinoma test images in which 1 or more malignant cells were detected was 2710 of 2731. The sensitivity in the nonadenocarcinoma dataset was 97.1%, and the false-positive rate in the negative-for-malignancy dataset was 7.3%. The accuracy, sensitivity, and specificity of the model using all the test datasets were 96.3%, 98.5%, and 92.7%, respectively.

Conclusions

Although some issues regarding cell annotation when creating an object detection model remain, the accuracy is sufficient to assist cancer screening in effusion cytology. It is vital to reliably detect malignant cells in effusion cytology, and the further development of automated systems to reduce false-negative results is expected.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningAdenocarcinomaCytodiagnosisNeural Networks, ComputerPleural Effusion, MalignantHumansSensitivity and SpecificityAlgorithms

Stichwörter der Autor:innen

YOLOartificial intelligencedeep learningeffusion cytologyobject detection

Autor:innen und Institutionen

  • Pathophysiology Sciences, Department of Integrated Health Sciences, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan.
  • Clinical Engineering, Faculty of Medical Sciences, Juntendo University, Urayasu, Chiba, Japan.
  • Department of Medical Technique, Division of Pathology, Nagoya University Hospital, Nagoya, Aichi, Japan.
  • Department of Pathology and Laboratory Medicine, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan.

ORCID-Profile: Katsuhide Ikeda

Förderung

  • Japan Society for the Promotion of Science

Zitieren

APA

Ikeda, K., Sakabe, N., Fukuda, K., Sato, S., Hara, T., Kobayashi, H., Nakaguro, M., Karube, K., & Nagata, K. (2025). Deep learning neural network of adenocarcinoma detection in effusion cytology. American Journal of Clinical Pathology. https://doi.org/10.1093/ajcp/aqaf067

Vancouver

Ikeda K, Sakabe N, Fukuda K, Sato S, Hara T, Kobayashi H, et al. Deep learning neural network of adenocarcinoma detection in effusion cytology. American Journal of Clinical Pathology. 2025. doi:10.1093/ajcp/aqaf067

BibTeX
@article{Ikeda2025,
  author = {Ikeda, Katsuhide and Sakabe, Nanako and Fukuda, Kenta and Sato, Shouichi and Hara, Toshiaki and Kobayashi, Harumi and Nakaguro, Masato and Karube, Kennosuke and Nagata, Kohzo},
  title = {Deep learning neural network of adenocarcinoma detection in effusion cytology},
  journal = {American Journal of Clinical Pathology},
  year = {2025},
  doi = {10.1093/ajcp/aqaf067}
}
Kostenlos starten →