[论文解读] Synthetic CT Generation from MRI using 3D Transformer-based Denoising Diffusion Model

Shaoyan Pan, Elham Abouei+11 more|arXiv (Cornell University)|May 31, 2023

Medical Imaging Techniques and Applications被引用 10

一句话总结

论文提出 MC-DDPM，一种基于 transformer 的去噪扩散模型，将 MRI 转换为用于放射治疗的高质量合成 CT，在分钟内在脑部和前列腺数据上取得出色结果。

ABSTRACT

Magnetic resonance imaging (MRI)-based synthetic computed tomography (sCT) simplifies radiation therapy treatment planning by eliminating the need for CT simulation and error-prone image registration, ultimately reducing patient radiation dose and setup uncertainty. We propose an MRI-to-CT transformer-based denoising diffusion probabilistic model (MC-DDPM) to transform MRI into high-quality sCT to facilitate radiation treatment planning. MC-DDPM implements diffusion processes with a shifted-window transformer network to generate sCT from MRI. The proposed model consists of two processes: a forward process which adds Gaussian noise to real CT scans, and a reverse process in which a shifted-window transformer V-net (Swin-Vnet) denoises the noisy CT scans conditioned on the MRI from the same patient to produce noise-free CT scans. With an optimally trained Swin-Vnet, the reverse diffusion process was used to generate sCT scans matching MRI anatomy. We evaluated the proposed method by generating sCT from MRI on a brain dataset and a prostate dataset. Qualitative evaluation was performed using the mean absolute error (MAE) of Hounsfield unit (HU), peak signal to noise ratio (PSNR), multi-scale Structure Similarity index (MS-SSIM) and normalized cross correlation (NCC) indexes between ground truth CTs and sCTs. MC-DDPM generated brain sCTs with state-of-the-art quantitative results with MAE 43.317 HU, PSNR 27.046 dB, SSIM 0.965, and NCC 0.983. For the prostate dataset, MC-DDPM achieved MAE 59.953 HU, PSNR 26.920 dB, SSIM 0.849, and NCC 0.948. In conclusion, we have developed and validated a novel approach for generating CT images from routine MRIs using a transformer-based DDPM. This model effectively captures the complex relationship between CT and MRI images, allowing for robust and high-quality synthetic CT (sCT) images to be generated in minutes.

研究动机与目标

• 通过 MRI 基于合成 CT 来减少 CT 扫描与图像配准误差，从而提升放射治疗的效率与准确性。
• 开发一种利用以患者 MRI 为条件的扩散概率模型将 MRI 转换为 CT 的方法。
• 在 DDPM 框架内引入一个 shift 窗格的转换器网络（Swin-Vnet），以在去噪 CT 的同时保留 MRI 的解剖结构。
• 在脑部和前列腺数据集上验证该方法，并给出定量的 CT 精度指标。

提出的方法

• 使用前向扩散过程向真实 CT 图像添加高斯噪声。
• 使用 Swin-Vnet 条件化去噪噪声 CT 图像的反向扩散过程，条件为患者的 MRI。
• 将三维基于转换器的去噪扩散概率模型（MC-DDPM）整合，以从 MRI 生成无噪声的 CT。
• 用 MAE、PSNR、MS-SSIM 以及 NCC 对生成的 sCT 与真值 CT 进行对比评估。
• 报告脑部与前列腔结果以体现数据集相关的性能差异。

实验结果

研究问题

• RQ1 MC-DDPM 能否在保持解剖保真度的同时从 MRI 精确生成 CT？
• RQ2 在脑部与前列腺数据集上，标准 sCT 指标的表现差异如何？
• RQ3 与真值 CT 相比，所提出方法在 MAE、PSNR、SSIM、NCC 上的定量提升是多少？
• RQ4 sCT 生成过程是否足够快速，适用于实际放射治疗计划（每例分钟级）？

主要发现

• 脑部 sCT：MAE 43.317 HU，PSNR 27.046 dB，SSIM 0.965，NCC 0.983。
• 前列腺 sCT：MAE 59.953 HU，PSNR 26.920 dB，SSIM 0.849，NCC 0.948。
• MC-DDPM 能够有效捕捉 CT–MRI 的关系，在数分钟内生成解剖结构匹配的 sCT。
• 基于 MRI 条件的 Swin-Vnet 在反向扩散过程中实现了对 CT 的高质量去噪。