A Generative Adversarial Network technique for high-quality super-resolution reconstruction of cardiac magnetic resonance images
A Generative Adversarial Network technique for high-quality super-resolution reconstruction of cardiac magnetic resonance images
复制标题
用于心脏磁共振图像高质量超分辨率重建的生成对抗网络技术
DOI：
10.1016/j.mri.2021.10.033
复制
发表时间：
2021-10-27
影响因子：
2.5
通讯作者：
Wong, Kelvin K. L.
中科院分区：
文献类型：
--
作者：
Zhao, Ming;Wei, Yang;Wong, Kelvin K. L.
用途：本文提出了一种去噪超分辨率生成对抗网络（DnSRGAN）方法，用于噪声心脏磁共振（CMR）图像的高质量超分辨率重建。方法：该方法基于前馈去噪卷积神经网络（DnCNN）和SRGAN架构。首先，我们使用前馈去噪神经网络对CMR图像进行预去噪，以确保输入是一幅干净的图像。其次，采用梯度惩罚（GP）方法解决了非线性梯度消失的问题，提高了模型的收敛速度。最后，在原有SRGAN损失函数的基础上增加新的损失函数，监测GAN梯度下降，实现更稳定、更高效的模型训练，从而为CMR图像的超分辨率提供更高的感知质量。结果：将所测试的心脏图像分为3组，每组25幅图像。然后，我们计算峰值信噪比（PSNR）/结构相似度（SSIM）之间的地面真实（GT）和超分辨率生成的图像，使用它们来评估我们的模型。与目前广泛使用的双三次ESRGAN和SRGAN方法相比，该方法具有更好的重建质量和更高的PSNR/SSIM评分。结论：采用DnCNN对CMR图像进行去噪处理，然后利用改进的SRGAN对去噪后的图像进行超分辨率重建，解决了超分辨率重建过程中高噪声和伪影导致心脏图像重建错误的问题。
Purpose: In this paper, we proposed a Denoising Super-resolution Generative Adversarial Network (DnSRGAN) method for high-quality super-resolution reconstruction of noisy cardiac magnetic resonance (CMR) images. Methods: The proposed method is based on feed-forward denoising convolutional neural network (DnCNN) and SRGAN architecture. Firstly, we used a feed-forward denoising neural network to pre-denoise the CMR image to ensure that the input is a clean image. Secondly, we use the gradient penalty (GP) method to solve the problem of the discriminator gradient disappearing, which improves the convergence speed of the model. Finally, a new loss function is added to the original SRGAN loss function to monitor GAN gradient descent to achieve more stable and efficient model training, thereby providing higher perceptual quality for the super-resolution of CMR images. Results: We divided the tested cardiac images into 3 groups, each group of 25 images. Then, we calculated the Peak Signal to Noise Ratio (PSNR) /Structural Similarity (SSIM) between Ground Truth (GT) and the images generated by super-resolution, used them to evaluate our model. We compared with the current widely used method: Bicubic ESRGAN and SRGAN, our method has better reconstruction quality and higher PSNR/SSIM score. Conclusion: We used DnCNN to denoise the CMR image, and then using the improved SRGAN to perform superresolution reconstruction of the denoised image, we can solve the problem of high noise and artifacts that cause the cardiac image to be reconstructed incorrectly during super-resolution.