MRI Super-Resolution with GAN and 3D Multi-Level DenseNet: Smaller, Faster, and Better
This work addresses the need for faster and more efficient MRI super-resolution in medical imaging, though it is incremental as it builds on existing deep learning and GAN approaches.
The paper tackled the problem of generating high-resolution MRI images from low-resolution inputs to improve clinical diagnosis, proposing a novel 3D CNN architecture called mDCSRN-GAN that outperformed other deep learning methods in quality and speed on a dataset of 1,113 subjects for 4x super-resolution.
High-resolution (HR) magnetic resonance imaging (MRI) provides detailed anatomical information that is critical for diagnosis in the clinical application. However, HR MRI typically comes at the cost of long scan time, small spatial coverage, and low signal-to-noise ratio (SNR). Recent studies showed that with a deep convolutional neural network (CNN), HR generic images could be recovered from low-resolution (LR) inputs via single image super-resolution (SISR) approaches. Additionally, previous works have shown that a deep 3D CNN can generate high-quality SR MRIs by using learned image priors. However, 3D CNN with deep structures, have a large number of parameters and are computationally expensive. In this paper, we propose a novel 3D CNN architecture, namely a multi-level densely connected super-resolution network (mDCSRN), which is light-weight, fast and accurate. We also show that with the generative adversarial network (GAN)-guided training, the mDCSRN-GAN provides appealing sharp SR images with rich texture details that are highly comparable with the referenced HR images. Our results from experiments on a large public dataset with 1,113 subjects showed that this new architecture outperformed other popular deep learning methods in recovering 4x resolution-downgraded images in both quality and speed.