Accelerating Diffusion Models Via Pre-Segmentation Diffusion Sampling for Medical Image Segmentation
Accelerating Diffusion Models Via Pre-Segmentation Diffusion Sampling for Medical Image Segmentation
复制标题
通过用于医学图像分割的预分割扩散采样加速扩散模型
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Tingxia Ma
中科院分区：
文献类型：
--
作者：
Xutao Guo;Yanwu Yang;Chenfei Ye;Shangfeng Lu;Yang Xiang;Tingxia Ma
基于去噪扩散概率模型（DDPM），医学图像分割可以被描述为条件图像生成任务，其允许计算分割的逐像素不确定性图，并且允许分割的隐式集合以提高分割性能。然而，DDPM需要许多迭代去噪步骤来从高斯噪声中生成分割，从而导致非常低效的推理。为了缓解这个问题，我们提出了一个原则性的加速策略，称为预分割扩散采样DDPM（PD-DDPM），这是专门用于医学图像分割。其核心思想是基于单独训练的分割网络获得预分割结果，并根据前向扩散规则构建噪声预测（非高斯分布）。然后，我们可以从噪声预测开始，使用更少的反向步骤来生成分割结果。实验表明，PD-DDPM产生更好的分割结果比有代表性的基线方法，即使反向步骤的数量显着减少。此外，PD-DDPM与现有的高级分割模型正交，可以将其组合以进一步提高分割性能。
Based on the Denoising Diffusion Probabilistic Model (DDPM), medical image segmentation can be described as a conditional image generation task, which allows to compute pixel-wise uncertainty maps of the segmentation and allows an implicit ensemble of segmentations to boost the segmentation performance. However, DDPM requires many iterative denoising steps to generate segmentations from Gaussian noise, resulting in extremely inefficient inference. To mitigate the issue, we propose a principled acceleration strategy, called pre-segmentation diffusion sampling DDPM (PD-DDPM), which is specially used for medical image segmentation. The key idea is to obtain pre-segmentation results based on a separately trained segmentation network, and construct noise predictions (non-Gaussian distribution) according to the forward diffusion rule. We can then start with noisy predictions and use fewer reverse steps to generate segmentation results. Experiments show that PD-DDPM yields better segmentation results over representative baseline methods even if the number of reverse steps is significantly reduced. Moreover, PD-DDPM is orthogonal to existing advanced segmentation models, which can be combined to further improve the segmentation performance.