Deep neural networks (DNNs) achieve promising performance in visual recognition under the independent and identically distributed (IID) hypothesis. In contrast, the IID hypothesis is not universally guaranteed in numerous real-world applications, especially in medical image analysis. Medical image segmentation is typically formulated as a pixel-wise classification task in which each pixel is classified into a category. However, this formulation ignores the hard-to-classified pixels, e.g., some pixels near the boundary area, as they usually confuse DNNs. In this paper, we first explore that hard-to-classified pixels are associated with high uncertainty. Based on this, we propose a novel framework that utilizes uncertainty estimation to highlight hard-to-classified pixels for DNNs, thereby improving its generalization. We evaluate our method on two popular benchmarks: prostate and fundus datasets. The results of the experiment demonstrate that our method outperforms state-of-the-art methods.
@article{arxiv.2305.07883,
title = {Towards Generalizable Medical Image Segmentation with Pixel-wise Uncertainty Estimation},
author = {Shuai Wang and Zipei Yan and Daoan Zhang and Zhongsen Li and Sirui Wu and Wenxuan Chen and Rui Li},
journal= {arXiv preprint arXiv:2305.07883},
year = {2023}
}