# Robust Cell-Level Classification for Liquid-Based Cervical Cytology Using Deep Transfer Learning: A Multi-Source Study Addressing Scanner-Induced Domain Shifts

## TL;DR

This study develops a robust deep learning model for classifying cervical cells in Pap smears, addressing challenges caused by differences in scanning equipment and laboratories.

## Contribution

A novel deep transfer learning framework that improves robustness to scanner-induced domain shifts in cervical cytology classification.

## Key findings

- ResNet50 achieved the highest accuracy (0.91) and macro-F1 score (0.91) among tested CNN models.
- Incorporating multi-center WSI data improved robustness to scanner-induced variations compared to public data alone.
- Combining diverse data sources helps mitigate domain shift in cell-level cervical cytology classification.

## Abstract

Automated analysis of liquid-based cervical cytology is increasingly supported by digital microscopy and deep learning. However, model generalization remains challenging due to scanner- and laboratory-induced domain shifts affecting color, texture, and morphology. In this study, we present a robust cell-level classification framework for liquid-based Pap smear cytology based on deep transfer learning, designed to operate under heterogeneous acquisition conditions. We construct a multi-source dataset by integrating three widely used public reference repositories (SIPaKMeD, Herlev, CRIC Cervix) with a proprietary cohort comprising 416 Whole Slide Images (WSIs) collected from two medical centers and digitized using different scanning systems. All labels are harmonized into four Bethesda categories (NILM, ASC-US, LSIL, HSIL), and cell-centered 224 × 224 patches are used as standardized inputs for model development and benchmarking. We evaluate state-of-the-art CNN backbones (ResNet50, EfficientNetB0, VGG16) and perform systematic ablation across data-source combinations to quantify robustness under acquisition variability. Among the evaluated models, ResNet50 yields the best overall performance on the independent test set (accuracy = 0.91; macro-F1 = 0.91), consistently outperforming EfficientNetB0 and VGG16. Importantly, incorporating proprietary multi-center WSI-derived data improves robustness to scanner-induced variation compared to training on public data alone. These findings demonstrate that combining diverse data sources can mitigate domain shift in cell-level cervical cytology classification. While clinically actionable screening requires slide-level aggregation (e.g., MIL-based WSI inference), the proposed classifier provides a robust component that can be integrated into end-to-end WSI screening pipelines in future work.