Active Label Cleaning for Reliable Detection of Electron Dense Deposits in Transmission Electron Microscopy Images

TL;DR

This paper introduces an active label cleaning approach that uses active learning and discrepancy analysis to denoise crowdsourced labels for electron dense deposit detection in microscopy images, significantly improving accuracy and reducing annotation costs.

Contribution

The proposed method combines active learning with discrepancy-based noise grading to efficiently clean crowdsourced labels, achieving high detection performance with limited expert annotation.

Findings

Achieves 67.18% AP50 on private dataset, 18.83% better than noisy labels

Reaches 95.79% of full expert annotation performance

Reduces annotation cost by 73.30%

Abstract

Automated detection of electron dense deposits (EDD) in glomerular disease is hindered by the scarcity of high-quality labeled data. While crowdsourcing reduces annotation cost, it introduces label noise. We propose an active label cleaning method to efficiently denoise crowdsourced datasets. Our approach uses active learning to select the most valuable noisy samples for expert re-annotation, building high-accuracy cleaning models. A Label Selection Module leverages discrepancies between crowdsourced labels and model predictions for both sample selection and instance-level noise grading. Experiments show our method achieves 67.18% AP\textsubscript{50} on a private dataset, an 18.83% improvement over training on noisy labels. This performance reaches 95.79% of that with full expert annotation while reducing annotation cost by 73.30%. The method provides a practical, cost-effective solution for developing reliable medical AI with limited expert resources.