CytoLens: a modular deep learning system for quantitative cervical cytology
- 1Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea
Abstract
Cervical cytology screening relies on the visual assessment of nuclear and cellular morphology across thousands of cells per slide, a labor-intensive task that could benefit from quantitative computational support. Here we present CytoLens, a modular deep learning system that detects and counts nuclei, delineates nuclei and whole cells, measures the nuclear-to-cell area ratio (N:C), classifies cells from multifocal images and scores candidate abnormality. We trained and tested each module on a public dataset matched to its task, without pooling datasets. CytoLens detected nuclei with an F1 score of 0.931 (95% confidence interval (CI), 0.914-0.946) and counted 75 of 100 fields to within one nucleus. Point-prompted segmentation achieved mean Dice coefficients of 0.901 for nuclei and 0.847 for whole cells and, without retraining, transferred to a second dataset in which measured N:C increased with lesion grade in parallel with expert outlines. Multifocal hierarchical classification reached a mean accuracy of 0.947 across four Bethesda-aligned categories, and abnormality scores ranked annotated abnormal candidates with an area under the receiver operating characteristic curve (AUROC) of 0.865. These results support the evaluation of CytoLens on patient-linked clinical specimens.
Related
- Weakly-Supervised Gated Attention Multiple Instance Learning for Multi-Center Cervical Cytology Slide Triage
- CERVEX: A Foundation-Model Framework With Built-In Explainable AI for Automated Cervical Cytology Classification From Pap Smear Images
- Prevalence and predictors of abnormal cervical cytology among women living with HIV and without HIV from the MACS/WIHS combined cohort study
- Cytology, High-Risk HPV Genotyping, and Histology in Cervical Pathology: A Retrospective Cohort Study.
- Combined plasma and liquid-based cytology microRNA signatures as biomarkers of high-grade cervical lesions: association with HPV infection and p16 expression.
BibTeX
@misc{lee2026cytolens,
title = {{CytoLens: a modular deep learning system for quantitative cervical cytology}},
author = {K. Lee and R. J. Yoon and R. S. Lee and D. Lee},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.13.26365471},
url = {https://www.medrxiv.org/content/10.64898/2026.10.13.26365471v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - CytoLens: a modular deep learning system for quantitative cervical cytology
AU - K. Lee
AU - R. J. Yoon
AU - R. S. Lee
AU - D. Lee
PY - 2026
DA - 2026/10/13
PB - medRxiv
DO - 10.64898/2026.10.13.26365471
UR - https://www.medrxiv.org/content/10.64898/2026.10.13.26365471v1
AB - Cervical cytology screening relies on the visual assessment of nuclear and cellular morphology across thousands of cells per slide, a labor-intensive task that could benefit from quantitative computational support. Here we present CytoLens, a modular deep learning system that detects and counts nuclei, delineates nuclei and whole cells, measures the nuclear-to-cell area ratio (N:C), classifies cells from multifocal images and scores candidate abnormality. We trained and tested each module on a public dataset matched to its task, without pooling datasets. CytoLens detected nuclei with an F1 score of 0.931 (95% confidence interval (CI), 0.914-0.946) and counted 75 of 100 fields to within one nucleus. Point-prompted segmentation achieved mean Dice coefficients of 0.901 for nuclei and 0.847 for whole cells and, without retraining, transferred to a second dataset in which measured N:C increased with lesion grade in parallel with expert outlines. Multifocal hierarchical classification reached a mean accuracy of 0.947 across four Bethesda-aligned categories, and abnormality scores ranked annotated abnormal candidates with an area under the receiver operating characteristic curve (AUROC) of 0.865. These results support the evaluation of CytoLens on patient-linked clinical specimens.
ER -