Superpixel-guided Iterative Learning from Noisy Labels for Medical Image Segmentation
This addresses the problem of noisy annotations in medical image segmentation, which is crucial for accurate diagnosis, but it is an incremental improvement over existing methods.
The paper tackles medical image segmentation with noisy labels by using superpixel-guided iterative learning to exploit structural constraints, achieving superior robustness and outperforming state-of-the-art methods on two benchmarks.
Learning segmentation from noisy labels is an important task for medical image analysis due to the difficulty in acquiring highquality annotations. Most existing methods neglect the pixel correlation and structural prior in segmentation, often producing noisy predictions around object boundaries. To address this, we adopt a superpixel representation and develop a robust iterative learning strategy that combines noise-aware training of segmentation network and noisy label refinement, both guided by the superpixels. This design enables us to exploit the structural constraints in segmentation labels and effectively mitigate the impact of label noise in learning. Experiments on two benchmarks show that our method outperforms recent state-of-the-art approaches, and achieves superior robustness in a wide range of label noises. Code is available at https://github.com/gaozhitong/SP_guided_Noisy_Label_Seg.