- Suggested Parameters for Clinical Infant Brain Imaging Using an Ultra-Low-Field Portable MRI System
A portable ultra-low-field MRI system (0.064T) offers a solution by enabling infant scans directly in the NICU, eliminating transport risks. The optimized protocol presented in this study achieves shorter, clinically feasible scan times using in-line processing, supporting a comprehensive imaging set. The optimized parameters showed marked improvement over default settings. However, whether deep learning reconstruction affects sensitivity to pediatric pathologies warrants evaluation in future prospective studies in diverse practice settings.
- MR Cranial Bone Imaging: Evaluation of Both Motion-Corrected and Automated Deep Learning Pseudo-CT Estimated MR Images
In this study, the authors developed automated motion correction and used deep learning to generate pseudo-CT cranial images from MR images. Compared with CT, pseudo-CT had 100% specificity and 100% sensitivity for suture closure and 100% specificity and 90% sensitivity for skull fractures.
Check out our Case Collections, Special Collections, and more. Read the latest AJNR updates
Improved Turnaround Times | Median time to first decision: 12 days