Project DetailsMagnetic Resonance Imaging (MRI) is a critical imaging modality in modern medicine, providing high-resolution images of soft tissues without the use of ionizing radiation. However, the high cost and complexity of traditional high-field MRI scanners have driven interest in low-field MRI systems. While more affordable and portable, low-field MRI suffers from several drawbacks, including lower signal-to-noise ratio (SNR), reduced resolution, and poorer contrast, which limits its clinical utility.
This work aims to address these challenges by developing a deep-learning-based reconstruction method that enhances low-field MRI scans using reference images from past high-field scans of the same subject. By using special loss functions and leveraging the high-quality features from the reference scans, our approach aims to significantly improve the quality of low-field MRI images, making them more viable for clinical use. The proposed method is validated using a series of experiments that demonstrate its effectiveness in reconstructing high-quality images from low-field MRI data, offering a promising solution for making MRI more accessible in resource-constrained settings.
