[论文解读] DenoMamba: A fused state-space model for low-dose CT denoising

Şaban Öztürk, Özge Duran+1 more|arXiv (Cornell University)|Sep 19, 2024

Medical Imaging Techniques and Applications被引用 7

一句话总结

DenoMamba 引入一种新颖的状态空间建模方法（FuseSSM 块）以融合空间上下文和通道上下文用于 LDCT 去噪，在与相似复杂度的 CNN/Transformer 基线相比达到更优图像质量。

ABSTRACT

Low-dose computed tomography (LDCT) lower potential risks linked to radiation exposure while relying on advanced denoising algorithms to maintain diagnostic quality in reconstructed images. The reigning paradigm in LDCT denoising is based on neural network models that learn data-driven image priors to separate noise evoked by dose reduction from underlying tissue signals. Naturally, the fidelity of these priors depend on the model's ability to capture the broad range of contextual features evident in CT images. Earlier convolutional neural networks (CNN) are highly adept at efficiently capturing short-range spatial context, but their limited receptive fields reduce sensitivity to interactions over longer distances. Although transformers based on self-attention mechanisms have recently been posed to increase sensitivity to long-range context, they can suffer from suboptimal performance and efficiency due to elevated model complexity, particularly for high-resolution CT images. For high-quality restoration of LDCT images, here we introduce DenoMamba, a novel denoising method based on state-space modeling (SSM), that efficiently captures short- and long-range context in medical images. Following an hourglass architecture with encoder-decoder stages, DenoMamba employs a spatial SSM module to encode spatial context and a novel channel SSM module equipped with a secondary gated convolution network to encode latent features of channel context at each stage. Feature maps from the two modules are then consolidated with low-level input features via a convolution fusion module (CFM). Comprehensive experiments on LDCT datasets with 25\% and 10\% dose reduction demonstrate that DenoMamba outperforms state-of-the-art denoisers with average improvements of 1.4dB PSNR, 1.1% SSIM, and 1.6% RMSE in recovered image quality.

研究动机与目标

• 在 LDCT 去噪中推动在辐射剂量降低与诊断图像质量之间取得平衡。
• 提出一种基于 SSM 的新去噪架构（DenoMamba），能够捕捉短程与长程上下文。
• 证明融合的时空通道 SSM 块在复杂度相当的情况下优于 CNN 和 Transformer。
• 在 LDCT 数据集上展示对 25% 和 10% 剂量降低的鲁棒去噪。
• 提供开源实现以促进可重复性。

提出的方法

• 引入将空间 SSM 模块与通道 SSM 模块结合的 FuseSSM 块。
• 在通道 SSM 中使用一个二级门控卷积网络来提取潜在通道特征。
• 通过卷积融合模块（CFM）将上下文表示与原始输入融合。
• 采用带有编码器和解码器阶段之间的长程残差连接的 hourglass 编码器-解码器架构。
• 使用 LDCT-to-NDCT 图像对的均方误差损失进行训练，并以 PSNR、SSIM 和 RMSE 进行评估。

实验结果

研究问题

• RQ1 在低剂量水平下，融合的时空通道状态空间模型是否能比 CNN/Transformer 基线改善 LDCT 去噪？
• RQ2 所提出的 DenoMamba 架构在有效捕获长程上下文的同时，是否能维持高空间精度？
• RQ3 移除组件（cha SSM、spa SSM、CFM、GCN、身份路径）对去噪性能的影响如何？
• RQ4 在 25% 和 10% 剂量降低及多种重建核下，DenoMamba 的表现如何？
• RQ5 该方法的代码是否可获得以便可重复性？

主要发现

• 在 25% 副剂量的 LDCT 数据集上，DenoMamba 比竞争方法获得更高的 PSNR 和 SSIM，且 RMSE 更低。
• 消融研究表明，空间 SSM 和通道 SSM 模块以及基于 GCN 的通道路径对于最佳性能都很重要。
• 所提出的方法在 1-mm 和 3-mm 切片厚度以及各种核下均展示出优越的图像质量。
• 该模型在参数数量与传统 CNN 相当的同时，能够以近似线性复杂度捕获长程上下文。
• DenoMamba 的代码公开可用。