CerS-Path: Self-supervised Diagnostic System for Cervical Pathology

It integrates eight diagnostic functions across biopsy to resection specimens and outperforms state-of-the-art foundation models in 26 of 30 tasks. In six independent cohorts, it achieves up to 99% sensitivity for clinically relevant lesions with acceptable specificity, demonstrating strong potential for screening-based histopathology workflows.

Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.