Self-supervised learning-based cervical cytology for the triage of HPV-positive women in resource-limited settings and low-data regime

Thomas Stegmüller, Christian Abbet, Behzad Bozorgtabar, Holly Clarke, Patrick Petignat, Pierre Vassilakos, Jean-Philippe Thiran

Abstract: Highlights•Medium-sized dataset of liquid-based cytology Pap smear test images.•Assessment of self-supervised learning for cytology images.•Data augmentation strategy tailored for cytology images.•Improved alignment of multiple instance learning (MIL) with Pap smear test images.