Low-Field Magnetic Resonance Image Quality Enhancement using Undersampled k-Space and Out-of-Distribution Generalisation
Computer Vision and Pattern Recognition
Summary
The authors developed a method to improve images from low-strength MRI machines, which are cheaper but usually slower and less clear. Their technique directly works on the raw data collected by the scanner to fill in missing parts and produce clearer brain images, similar to those from stronger MRI machines. They also tested how well their method works on new data that is different from what it was trained on, and included a way to show how confident the model is in its results. Their approach performed better than existing methods that work on processed images, making it a promising way to enhance low-field MRI scans.
Authors
Daniel Tweneboah Anyimadu, Mohammed M. Abdelsamea, Ahmed Karam Eldaly
Abstract
Low-field magnetic resonance imaging (MRI) offers affordable access to diagnostic imaging but faces challenges such as prolonged acquisition times and reduced image quality. Although accelerated imaging via k-space undersampling helps reduce scan time, image quality enhancement methods often rely on spatial-domain postprocessing. Deep learning achieved state-of-the-art results in both domains. However, most models are trained and evaluated using in-distribution (InD) data, creating a significant gap in understanding model performance when tested using out-of-distribution (OOD) data. To address these issues, we propose a novel framework that reconstructs high-field-like MR images directly from undersampled low-field MRI k-space, quantifies the impact of reduced sampling, and evaluates the generalisability of the model using OOD. Our approach utilises a k-space dual channel U-Net to jointly process the real and imaginary components of undersampled k-space, restoring missing frequency content, and incorporates an ensemble strategy to generate uncertainty maps. Experiments on low-field brain MRI demonstrate that our k-space-driven image quality enhancement outperforms the counterpart spatial-domain and other state-of-the-art baselines, achieving image quality comparable to full high-field k-space acquisitions using OOD data. To the best of our knowledge, this work is among the first to combine low-field MR image reconstruction, quality enhancement using undersampled k-space, and uncertainty quantification within a unified framework.