3D High-Quality Magnetic Resonance Image Restoration in Clinics Using Deep Learning

Shortening acquisition time and reducing the motion artifacts are two of the most essential concerns in magnetic resonance imaging. As a promising solution, deep learning-based high-quality MR image restoration has been investigated to generate highly-resolved and motion artifact-free MR images from lower resolution images acquired with shortened acquisition time or motion artifact-corrupted images. However, numerous problems still exist to prevent deep learning approaches from becoming practical in the clinic environment. Specifically, most of the prior works focus solely on the network but ignore the impact of various down-sampling strategies on the acquisition time. Besides, the long inference time and high GPU consumption are also the bottlenecks to deploy most of the prior works in clinics. Furthermore, prior studies employ random movement in retrospective motion artifact generation, resulting in uncontrollable severity of motion artifact. More importantly, doctors are unsure whether the generated MR images are trustworthy, making diagnosis difficult. To overcome all these problems, we adopted a unified framework of 2D deep learning neural network for both 3D MRI super-resolution and motion artifact reduction, demonstrating such a framework can achieve better performance in 3D MRI restoration tasks compared to other state-of-the-art methods and remain the GPU consumption and inference time significantly low, thus easier to deploy. We also analyzed several down-sampling strategies based on the acceleration factor, including multiple combinations of in-plane and through-plane down-sampling, and developed a controllable and quantifiable motion artifact generation method. At last, the pixel-wise uncertainty was calculated and used to estimate the accuracy of the generated image, providing additional information for a reliable diagnosis.