Active Label Cleaning for Reliable Detection of Electron Dense Deposits in Transmission Electron Microscopy Images
This provides a cost-effective solution for developing reliable medical AI in domains like glomerular disease diagnosis, where expert annotation is scarce, though it is incremental as it builds on existing active learning and label cleaning techniques.
The paper tackles the problem of label noise in crowdsourced datasets for detecting electron dense deposits in medical images by proposing an active label cleaning method, which improves detection accuracy by 18.83% in AP50 and reduces annotation costs by 73.30% compared to using noisy labels.
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.