Adaptive Gradient Balancing for Undersampled MRI Reconstruction and Image-to-Image Translation
This work addresses the challenge of preserving fine details in accelerated MRI scans, which is incremental as it builds on existing deep learning approaches to improve image quality.
The paper tackles the problem of reconstructing sharp, high-quality images from undersampled MRI data by introducing an Adaptive Gradient Balancing technique and a Densely Connected Iterative Network, resulting in minimized artifacts and sharper images compared to other methods, with further validation on image-to-image translation tasks.
Recent accelerated MRI reconstruction models have used Deep Neural Networks (DNNs) to reconstruct relatively high-quality images from highly undersampled k-space data, enabling much faster MRI scanning. However, these techniques sometimes struggle to reconstruct sharp images that preserve fine detail while maintaining a natural appearance. In this work, we enhance the image quality by using a Conditional Wasserstein Generative Adversarial Network combined with a novel Adaptive Gradient Balancing (AGB) technique that automates the process of combining the adversarial and pixel-wise terms and streamlines hyperparameter tuning. In addition, we introduce a Densely Connected Iterative Network, which is an undersampled MRI reconstruction network that utilizes dense connections. In MRI, our method minimizes artifacts, while maintaining a high-quality reconstruction that produces sharper images than other techniques. To demonstrate the general nature of our method, it is further evaluated on a battery of image-to-image translation experiments, demonstrating an ability to recover from sub-optimal weighting in multi-term adversarial training.