cs.CVNov 3, 2026

When Cell Crops Share a Parent: Measuring and Isolating Acquisition Leakage in Cervical Cytology Benchmarks

Authors: Snehasish Ghosh, Soham Mitra, Antareep Majumdar, KC Santosh, Nibaran Das

Organizations: Independent Researcher, formerly affiliated with the Indian Institute of Science, Bengaluru, India · Department of Computer Science and Engineering, Jadavpur University, Kolkata, India · Qusbi Infotech Private Limited, Kolkata, India · Oregon Institute of Technology, USA

Abstract

Public cervical cytology benchmarks release many cell crops cut from far fewer microscope fields or slides and are often split at the crop level, so test cells share acquisition parents with training cells. In a preregistered study, crop-level splitting raised SIPaKMeD accuracy by 0.97 percentage points across twelve architectures and twenty seeds (95% CI 0.59-1.39), and CRIC accuracy by 12.08 points (11.16-12.99). On SIPaKMeD, a factorial that fixes training crop and parent counts attributes the gain to sibling exposure (1.73 points, 0.95-2.51), not to the extra parent diversity of crop-level splits (0.00). Parent identity is decodable at about 100 times chance, yet erasing 82% of its above-floor linear recoverability left the gap unchanged, and Grad-CAM did not distinguish leaking from clean models; matching test cells' color statistics to an unseen same-class field removed most of the advantage. Decodability and saliency cannot substitute for a parent-disjoint split.

Figures & tables

Explore similar work

CardsList
  1. Deep Learning for Cervical Cytology Screening: Methods, Algorithms, Challenges and Prospects

    Nov 1, 2026Jiang Yangbo, Zheng NengganMedical Image ClassificationComputational Pathology

  2. Reliability-Aware Hybrid-K Ensemble Selection for Cervical Cytology Classification: Integrating Discrimination, Calibration, and Selective Prediction

    Sep 1, 2026Nisreen Albzour, Sarah S. LamComputational PathologyEnsemble Learning