Pathological Image Segmentation with Noisy Labels
This addresses inconsistent manual labels from pathologists in medical image segmentation, which is crucial for disease diagnosis.
The paper tackles the problem of pathological image segmentation with noisy labels by proposing a label re-weighting framework and attention heatmap method, achieving improved robustness and outperforming state-of-the-art approaches on the Gleason 2019 datasets.
Segmentation of pathological images is essential for accurate disease diagnosis. The quality of manual labels plays a critical role in segmentation accuracy; yet, in practice, the labels between pathologists could be inconsistent, thus confusing the training process. In this work, we propose a novel label re-weighting framework to account for the reliability of different experts' labels on each pixel according to its surrounding features. We further devise a new attention heatmap, which takes roughness as prior knowledge to guide the model to focus on important regions. Our approach is evaluated on the public Gleason 2019 datasets. The results show that our approach effectively improves the model's robustness against noisy labels and outperforms state-of-the-art approaches.