Invention Reference Number
Ultra-high field MRI provides exceptional resolution but suffers from uneven magnetic field distribution, leading to image artifacts and reduced diagnostic reliability. This invention introduces a learning-based framework that dramatically accelerates and improves RF shimming, enabling more consistent image quality. By addressing one of the major technical barriers to ultra-high field MRI, the technology enhances both research and clinical imaging workflows.
Description
RF shimming is critical for balancing magnetic field uniformity in advanced MRI systems, especially at ultra-high field strengths. Traditional optimization approaches often require extensive time and patient-specific calibration, limiting clinical adoption. The Fast-RF-Shimming framework leverages a deep learning approach that reduces computation time from minutes to fractions of a second while maintaining or improving field homogeneity. The system incorporates predictive modeling to determine shimming parameters and includes an optional quality-control step to detect irregular outcomes. This design minimizes the need for on-site computational resources and enables near real-time adjustments, creating opportunities for broader use of ultra-high field imaging in both medical and research applications.
Benefits
- Rapid shimming with dramatically reduced computation times
- Improved imaging consistency and reliability
- Optional safeguard to detect and reduce non-uniform artifacts
Applications and Industries
- Medical imaging for neurology, cardiology, and oncology
- Research institutions advancing MRI-based studies
- Manufacturers of MRI systems and imaging hardware
Contact
To learn more about this technology, email partnerships@ornl.gov or call 865-574-1051.
