In this challenge, participants are invited to develop models for cell detection and classification in Pap smear images, with the goal of advancing automated decision-support tools for cervical cancer screening. The challenge is based on the RIVA Dataset, a high-quality open dataset with thousands of independent expert annotations, designed as a benchmark for cell detection and classification in conventional cytology.
CHALLENGE TRACKSAWARDS- Track A: Cell detection and classification into 8 diagnostic categories (Bethesda system).- Track B: Cell detection across diagnostic categories (without classification).
- Track A: USD 2,000 (1st place) · USD 1,200 (2nd place).
- Track B: USD 1,200 (1st place) · USD 600 (2nd place).IMPORTANT DATES
- Dec 15, 2025 : Official launch & preliminary test set release.PAPER SUBMISSION & ATTENDANCE- Feb 13, 2026 : End of preliminary phase / start of final evaluation.- Feb 20, 2026 : End of final evaluation.- Feb 26, 2026 : Paper submission deadline (IEEE ISBI format).- Apr 8–11, 2026: Awards ceremony at IEEE ISBI 2026.
All information (rules, timeline, evaluation metrics, and resources) is available at: https://lia-ditella.github.io/rivachallenge/Best,
RIVA Challenge Organizing Team