Chan Karissa, Maralani Pejman Jabehdar, Moody Alan R, Khademi April
Electrical, Computer and Biomedical Engineering Department, Toronto Metropolitan University, Toronto, ON, Canada.
Institute of Biomedical Engineering, Science and Technology (iBEST), Toronto, ON, Canada.
Front Neuroinform. 2023 Aug 2;17:1197330. doi: 10.3389/fninf.2023.1197330. eCollection 2023.
Acquisition and pre-processing pipelines for diffusion-weighted imaging (DWI) volumes are resource- and time-consuming. Generating synthetic DWI scalar maps from commonly acquired brain MRI sequences such as fluid-attenuated inversion recovery (FLAIR) could be useful for supplementing datasets. In this work we design and compare GAN-based image translation models for generating DWI scalar maps from FLAIR MRI for the first time.
We evaluate a pix2pix model, two modified CycleGANs using paired and unpaired data, and a convolutional autoencoder in synthesizing DWI fractional anisotropy (FA) and mean diffusivity (MD) from whole FLAIR volumes. In total, 420 FLAIR and DWI volumes (11,957 images) from multi-center dementia and vascular disease cohorts were used for training/testing. Generated images were evaluated using two groups of metrics: (1) human perception metrics including peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), (2) structural metrics including a newly proposed histogram similarity (Hist-KL) metric and mean squared error (MSE).
Pix2pix demonstrated the best performance both quantitatively and qualitatively with mean PSNR, SSIM, and MSE metrics of 23.41 dB, 0.8, 0.004, respectively for MD generation, and 24.05 dB, 0.78, 0.004, respectively for FA generation. The new histogram similarity metric demonstrated sensitivity to differences in fine details between generated and real images with mean pix2pix MD and FA Hist-KL metrics of 11.73 and 3.74, respectively. Detailed analysis of clinically relevant regions of white matter (WM) and gray matter (GM) in the pix2pix images also showed strong significant ( < 0.001) correlations between real and synthetic FA values in both tissue types ( = 0.714 for GM, = 0.877 for WM).
DISCUSSION/CONCLUSION: Our results show that pix2pix's FA and MD models had significantly better structural similarity of tissue structures and fine details than other models, including WM tracts and CSF spaces, between real and generated images. Regional analysis of synthetic volumes showed that synthetic DWI images can not only be used to supplement clinical datasets, but demonstrates potential utility in bypassing or correcting registration in data pre-processing.
扩散加权成像（DWI）容积的采集和预处理流程资源消耗大且耗时。从常用的脑部MRI序列（如液体衰减反转恢复序列（FLAIR））生成合成DWI标量图可能有助于补充数据集。在这项工作中，我们首次设计并比较了基于生成对抗网络（GAN）的图像转换模型，用于从FLAIR MRI生成DWI标量图。
我们评估了一个pix2pix模型、两个使用配对和未配对数据的改进型CycleGAN以及一个卷积自动编码器，用于从完整的FLAIR容积中合成DWI分数各向异性（FA）和平均扩散率（MD）。总共使用了来自多中心痴呆和血管疾病队列的420个FLAIR和DWI容积（11957张图像）进行训练/测试。使用两组指标评估生成的图像：（1）人类感知指标，包括峰值信噪比（PSNR）和结构相似性（SSIM）；（2）结构指标，包括新提出的直方图相似性（Hist-KL）指标和均方误差（MSE）。
Pix2pix在定量和定性方面均表现出最佳性能，生成MD时的平均PSNR、SSIM和MSE指标分别为23.41 dB、0.8、0.004，生成FA时分别为24.05 dB、0.78、0.004。新的直方图相似性指标对生成图像和真实图像之间的精细细节差异具有敏感性，pix2pix生成MD和FA的平均Hist-KL指标分别为11.73和3.74。对pix2pix图像中白质（WM）和灰质（GM）的临床相关区域进行的详细分析还表明，两种组织类型中真实和合成FA值之间存在强显著（<0.001）相关性（GM为0.714，WM为0.877）。
讨论/结论：我们的结果表明，pix2pix的FA和MD模型在真实图像和生成图像之间的组织结构和精细细节的结构相似性方面明显优于其他模型，包括WM束和脑脊液间隙。对合成容积的区域分析表明，合成DWI图像不仅可用于补充临床数据集，还在数据预处理中绕过或校正配准方面显示出潜在效用。
