Towards Global AI-Driven Cervical Cancer Screening

2026-06-12

AI Summary

This study addresses the limited accessibility of cervical cancer screening in low- and middle-income countries due to shortages of expert pathologists and biopsy resources. To overcome this challenge, the authors propose a multi-task deep learning approach that simultaneously performs lesion segmentation and image-level classification on colposcopic images, enhanced by large-scale data augmentation to account for visual heterogeneity. The method is the first to be validated on real-world, multi-country data, moving beyond prior reliance on single-country private datasets and substantially improving global generalizability. In internal validation, the model achieved a balanced accuracy of 0.68, outperforming medical experts (0.64); it also surpassed baseline methods across external test sets from four countries, with AUC scores ranging from 0.54 to 0.80.

Abstract

The global elimination of cervical cancer is a key public health goal set by the World Health Organization (WHO), with screening programs reducing mortality by up to 80%. However, access to experts and biopsy services is limited in low- to middle-income countries (LMICs). Deep learning (DL)-based algorithms offer promising support for screening, but most existing approaches have been developed and validated on private datasets from single countries. We present the first DL-based approach to cervical cancer screening validated on data from multiple countries. Technically, we phrase the problem of detecting and classifying lesions in colposcopy images as a multi-task learning problem, in which we simultaneously perform image-level classification and lesion segmentation. Our model was trained on a private data set of acid stain colposcopy images with manually generated lesion segmentation masks and corresponding histopathological results, employing extensive data augmentation to address image variability. In an in-distribution validation with pathology results serving as ground truth, our algorithm outperformed medical experts (Balanced Accuracy: 0.68 vs 0.64) in CIN1- (Cervical intraepithelial neoplasia grade 1 or lower) versus CIN2+ (grade 2 or higher) classification. External validation on four colposcopy data sets from four countries featuring radical differences in prevalence and patient characteristics yielded superior performance of our method compared to baseline methods. Performance variability across countries was high with AUC values ranging from 0.54 - 0.80. Overall, algorithm performance varied with age, transformation zone (cervical area most prone to lesion development), presence of comorbidities and pathognomonic signs, with comorbidities having by far the largest negative effect. Future work should focus on improving model robustness and generalizability.

Problem

• cervical cancer screening
• deep learning
• multi-country validation
• generalizability
• colposcopy

Innovation

• multi-task learning
• cross-country validation
• colposcopy image analysis
• lesion segmentation
• AI-driven screening