Weakly supervised nuclei segmentation based on pseudo label correction and uncertainty denoising

Xipeng Pan, Shilong Song, Zhenbing Liu, Huadeng Wang, Lingqiao Li, Haoxiang Lu, Rushi Lan, Xiaonan Luo

Abstract: Highlights•We propose a weakly supervised segmentation framework using only point annotations.•EMAC and uncertainty denoising are proposed to improve the quality of pseudo label.•Adaptive clustering method is designed to adapt the color distribution of different images.•Our proposed method is superior to other existing nuclei segmentation methods.