EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Artificial Intelligence-Assisted Colposcopy: Deep Learning Multi-Class Segmentation of Anatomical Structures and Pathological Findings for Cervical Cancer Screening

Marcin Jurczak, Łukasz Charzewski, Beata Goźlińska, Paweł Albrycht iD, Kacper Kobus, Artur Ludwin iD, Zoulikha Jabiry-Zieniewicz, Sylwester Kominek, Grzegorz Basiński, Bartosz Korzeb, Barbara Ewa Suchońska iD

Cancers · 05.05.2026 · 18 (9) · 1485

DOI: 10.3390/cancers18091485 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cancers, online erschienen am 05.05.2026. Verfasst von Marcin Jurczak und 10 weiteren Autor:innen. Beteiligte Einrichtung: Proacta SA, Srebrna 16, 00-810 Warsaw, Poland. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Medical Research Agency, Poland.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate colposcopy assessment is essential for detection of cervical precancerous lesions. However, the diagnostic performance depends heavily on the examiner’s experience and subjective interpretation. Recent advances in artificial intelligence offer new opportunities for automated image analysis. In particular, deep learning models have shown promise for colposcopy image analysis intended for precancer screening. However, their limited interpretability restricts their translational value in clinical gynecology. This study leverages a custom dataset of expert-annotated digital colposcopic images to quantify the diagnostic strengths of these architectures.

Methods

A comparative analysis was provided of the well-established convolutional neural network YOLOv11 and the modern transformer-based RF-DETR architecture, both of which were trained for the segmentation of 10 distinct classes. Target objects included anatomical structures, medical instruments and colposcopic findings.

Results

Our results demonstrate that the YOLO architecture provides better performance for anatomical structures, whereas the RF-DERT reaches higher scores for colposcopy findings. These findings demonstrate transformer architecture superiority in more nuanced segmentation of clinically relevant findings, providing an interesting framework for computer-aided decision support systems in colposcopy.

Conclusions

Both YOLOv11 and RF-DETR enable effective segmentation of colposcopic images, with performance dependent on class size and characteristics. Best results were achieved for large anatomical structures, while small and underrepresented findings remained challenging due to class imbalance. YOLO offers greater stability and efficiency, whereas RF-DETR performs better on more complex cases. Limitations include data imbalance, variable image quality, and annotation inconsistencies; future work should address these issues to improve generalization.

Schlagwörter

Stichwörter der Autor:innen

AIcolposcopycomputer visiondeep learningprecancer screeningsemantic segmentation

Autor:innen und Institutionen

  • Proacta SA, Srebrna 16, 00-810 Warsaw, Poland.
  • University Center for Women's and Newborn's Health, Medical University of Warsaw, 1/3 Starynkiewicza Sq, 02-015 Warsaw, Poland.
  • 1st Department of Obstetrics and Gynaecology, Medical University of Warsaw, 1/3 Starynkiewicza Sq, 02-015 Warsaw, Poland.

ORCID-Profile: Paweł Albrycht, Artur Ludwin, Barbara Ewa Suchońska

Förderung

  • The Medical Research Agency, Poland

Zitieren

APA

Jurczak, M., Charzewski, Ł., Goźlińska, B., Albrycht, P., Kobus, K., Ludwin, A., Jabiry-Zieniewicz, Z., Kominek, S., Basiński, G., Korzeb, B., & Suchońska, B. E. (2026). Artificial Intelligence-Assisted Colposcopy: Deep Learning Multi-Class Segmentation of Anatomical Structures and Pathological Findings for Cervical Cancer Screening. Cancers, 18(9), 1485. https://doi.org/10.3390/cancers18091485

Vancouver

Jurczak M, Charzewski Ł, Goźlińska B, Albrycht P, Kobus K, Ludwin A, et al. Artificial Intelligence-Assisted Colposcopy: Deep Learning Multi-Class Segmentation of Anatomical Structures and Pathological Findings for Cervical Cancer Screening. Cancers. 2026;18(9):1485. doi:10.3390/cancers18091485

BibTeX
@article{Jurczak2026,
  author = {Jurczak, Marcin and Charzewski, Łukasz and Goźlińska, Beata and Albrycht, Paweł and Kobus, Kacper and Ludwin, Artur and Jabiry-Zieniewicz, Zoulikha and Kominek, Sylwester and Basiński, Grzegorz and Korzeb, Bartosz and Suchońska, Barbara Ewa},
  title = {Artificial Intelligence-Assisted Colposcopy: Deep Learning Multi-Class Segmentation of Anatomical Structures and Pathological Findings for Cervical Cancer Screening},
  journal = {Cancers},
  year = {2026},
  volume = {18},
  number = {9},
  pages = {1485},
  doi = {10.3390/cancers18091485}
}
Kostenlos starten →