Magnitude-image based data-consistent deep learning method for MRI super resolution
This work addresses MRI scan time reduction for clinical diagnosis by enhancing super-resolution image quality, but it is incremental as it builds on existing data consistency techniques.
The paper tackles the problem of artifacts in deep learning-based MRI super-resolution caused by training-testing data discrepancies, proposing a magnitude-image based data-consistent method that improves image quality without requiring raw k-space data. The result shows improvements in NRMSE and SSIM compared to a baseline CNN without the data consistency module.
Magnetic Resonance Imaging (MRI) is important in clinic to produce high resolution images for diagnosis, but its acquisition time is long for high resolution images. Deep learning based MRI super resolution methods can reduce scan time without complicated sequence programming, but may create additional artifacts due to the discrepancy between training data and testing data. Data consistency layer can improve the deep learning results but needs raw k-space data. In this work, we propose a magnitude-image based data consistency deep learning MRI super resolution method to improve super resolution images' quality without raw k-space data. Our experiments show that the proposed method can improve NRMSE and SSIM of super resolution images compared to the same Convolutional Neural Network (CNN) block without data consistency module.