Noisy labels can significantly affect the performance of deep neural networks (DNNs). In medical image segmentation tasks, annotations are error-prone due to the high demand in annotation time and in the annotators' expertise. Existing methods mostly assume noisy labels in different pixels are \textit{i.i.d}. However, segmentation label noise usually has strong spatial correlation and has prominent bias in distribution. In this paper, we propose a novel Markov model for segmentation noisy annotations that encodes both spatial correlation and bias. Further, to mitigate such label noise, we propose a label correction method to recover true label progressively. We provide theoretical guarantees of the correctness of the proposed method. Experiments show that our approach outperforms current state-of-the-art methods on both synthetic and real-world noisy annotations.
@article{arxiv.2308.02498,
title = {Learning to Segment from Noisy Annotations: A Spatial Correction Approach},
author = {Jiachen Yao and Yikai Zhang and Songzhu Zheng and Mayank Goswami and Prateek Prasanna and Chao Chen},
journal= {arXiv preprint arXiv:2308.02498},
year = {2023}
}