中文渐进式$\mathcal{J}$-不变自监督学习用于低剂量CT去噪
ENProgressive $\mathcal{J}$-Invariant Self-supervised Learning for Low-Dose CT Denoising
自监督学习用于低剂量CT（LDCT）去噪可减少对配对正常剂量数据的依赖，但现有盲点方法因感受野受限导致训练效率低、性能次优。本文提出渐进式J-invariant学习，通过最大化利用J-invariant特性增强LDCT去噪，提升效果与效率。
arXiv:2601.14180v4 Announce Type: replace Abstract: Self-supervised learning has been increasingly investigated for low-dose computed tomography (LDCT) image denoising, as it alleviates the dependence on paired normal-dose CT (NDCT) data, which are often difficult to collect. However, many existing self-supervised blind-spot denoising methods suffer from training inefficiencies and suboptimal performance due to restricted receptive fields. To mitigate this issue, we propose a novel Progressive $\mathcal{J}$-invariant Learning that maximizes the use of $\mathcal{J}$-invariant to enhance LDCT de