Shanghai Institute for Advanced Communication and Data Science, Shanghai University, 200444 Shanghai, People's Republic of China. School of Communication and Information Engineering, Shanghai University, 200444 Shanghai, People's Republic of China.
Phys Med Biol. 2018 Apr 19;63(8):085011. doi: 10.1088/1361-6560/aab9e9.
Spatial resolution is one of the key parameters of magnetic resonance imaging (MRI). The image super-resolution (SR) technique offers an alternative approach to improve the spatial resolution of MRI due to its simplicity. Convolutional neural networks (CNN)-based SR algorithms have achieved state-of-the-art performance, in which the global residual learning (GRL) strategy is now commonly used due to its effectiveness for learning image details for SR. However, the partial loss of image details usually happens in a very deep network due to the degradation problem. In this work, we propose a novel residual learning-based SR algorithm for MRI, which combines both multi-scale GRL and shallow network block-based local residual learning (LRL). The proposed LRL module works effectively in capturing high-frequency details by learning local residuals. One simulated MRI dataset and two real MRI datasets have been used to evaluate our algorithm. The experimental results show that the proposed SR algorithm achieves superior performance to all of the other compared CNN-based SR algorithms in this work.
空间分辨率是磁共振成像 (MRI) 的关键参数之一。由于其简单性，图像超分辨率 (SR) 技术为提高 MRI 的空间分辨率提供了一种替代方法。基于卷积神经网络 (CNN) 的 SR 算法已经取得了最先进的性能，其中全局残差学习 (GRL) 策略由于其对学习 SR 图像细节的有效性而被广泛使用。然而，由于退化问题，在非常深的网络中通常会部分丢失图像细节。在这项工作中，我们提出了一种新的基于残差学习的 MRI SR 算法，该算法结合了多尺度 GRL 和基于浅层网络块的局部残差学习 (LRL)。所提出的 LRL 模块通过学习局部残差，有效地捕获高频细节。使用一个模拟 MRI 数据集和两个真实 MRI 数据集来评估我们的算法。实验结果表明，所提出的 SR 算法在这项工作中优于所有其他基于 CNN 的比较 SR 算法。
