改进差分隐私的医疗图像分割方法，提升隐私保护下的分割精度。
From Gradient Clipping to Structural Refinement: Improving DPSGD for Medical Image Segmentation
- 结合梯度裁剪与形态学优化，提升隐私保护下的分割效果。
- 实验表明，现有方法在医学图像上表现不稳定，需针对性调整。
- 新模型自适应捕捉类别结构，适合多类医学图像分割任务。
医疗图像分割广泛用于疾病检测，但依赖敏感数据，存在隐私泄露风险。差分隐私通过差分私有随机梯度下降（DPSGD）提供保护，但会降低模型效用。现有DPSGD变体如自动裁剪（Auto-S）、归一化扰动SGD（NSGD）和逐样本自适应裁剪（PSAC）在图像分类中表现良好，但在医学分割中尚不明确。本文在二分类与多分类任务上评估这些方法，分析梯度对齐情况，发现此前假设（尤其是针对PSAC）并不稳定。进一步表明，将裁剪策略与形态学细化结合可提升隐私约束下的分割质量。最后提出自适应DP-Morph方法，能捕获类别特异性结构，在多分类场景中显著提升性能。
原文摘要 · Abstract (English)
Medical image segmentation is widely used for disease detection but relies on sensitive data, raising privacy concerns as trained models can leak information. Differential privacy, typically implemented via Differential Private Stochastic Gradient Descent (DPSGD), provides a solution, though at the cost of reduced utility. Recent DPSGD variants, including Automatic clipping (Auto-S), Normalised SGD with perturbation (NSGD), and Per-sample adaptive clipping (PSAC), have shown promise in image classification, but their behavior in medical segmentation remains underexplored. We evaluate these methods across binary and multi-class tasks and analyze gradient alignment, showing that prior assumptions, particularly for PSAC, do not consistently hold. We further demonstrate that combining clipping strategies with morphological refinement improves segmentation quality under privacy constraints. Finally, we propose an adaptive DP-Morph variant that captures class-specific structures and enhances performance in multi-class settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。