用潜在空间扩散模型实现高效且多解的医学图像分割。
Diffusion Model in Latent Space for Medical Image Segmentation Task
- 先用VAE压缩图像到低维潜在空间，再在该空间运行扩散模型。
- 在ISIC-2018等数据集上达到最优或接近最优的Dice和IoU分数。
- 可生成多个分割结果与置信图，适合临床决策支持。
医学图像分割对临床诊断与治疗规划至关重要。传统方法仅输出单一分割掩码，无法体现内在不确定性。近期生成模型能为每张图像生成多个合理掩码，模拟多位医生的协作判断，但计算开销大。本文提出MedSegLatDiff，一种基于扩散模型的框架，结合变分自编码器（VAE）与潜在扩散模型，实现高效分割。VAE将输入压缩至低维潜在空间，降低噪声并加速训练，扩散过程直接在该紧凑表示中进行。进一步在VAE掩码重建路径中采用加权交叉熵损失，更好保留小病灶等微小结构。在ISIC-2018（皮肤病变）、CVC-Clinic（息肉）和LIDC-IDRI（肺结节）数据集上评估，该模型在保持先进性能的同时，生成多样化的分割假设与置信图，显著提升可解释性与可靠性，适用于临床部署。
原文摘要 · Abstract (English)
Medical image segmentation is crucial for clinical diagnosis and treatment planning. Traditional methods typically produce a single segmentation mask, failing to capture inherent uncertainty. Recent generative models enable the creation of multiple plausible masks per image, mimicking the collaborative interpretation of several clinicians. However, these approaches remain computationally heavy. We propose MedSegLatDiff, a diffusion based framework that combines a variational autoencoder (VAE) with a latent diffusion model for efficient medical image segmentation. The VAE compresses the input into a low dimensional latent space, reducing noise and accelerating training, while the diffusion process operates directly in this compact representation. We further replace the conventional MSE loss with weighted cross entropy in the VAE mask reconstruction path to better preserve tiny structures such as small nodules. MedSegLatDiff is evaluated on ISIC-2018 (skin lesions), CVC-Clinic (polyps), and LIDC-IDRI (lung nodules). It achieves state of the art or highly competitive Dice and IoU scores while simultaneously generating diverse segmentation hypotheses and confidence maps. This provides enhanced interpretability and reliability compared to deterministic baselines, making the model particularly suitable for clinical deployment.
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。