Publications by authors named "K Ugurbil"
Purpose: to propose a two-step non-local principal component analysis (PCA) method and demonstrate its utility for denoising diffusion tensor MRI (DTI) with a few diffusion directions.
Methods: A two-step denoising pipeline was implemented to ensure accurate patch selection even with high noise levels and was coupled with data preprocessing for g-factor normalization and phase stabilization before data denoising with a non-local PCA algorithm. At the heart of our proposed pipeline was the use of a data-driven optimal shrinkage algorithm to manipulate the singular values in a way that would optimally estimate the noise-free signal.
Purpose: To develop and characterize the performance of a 128-channel head array for brain imaging at 10.5 tesla and evaluate the potential of brain imaging at this unique, >10 tesla magnetic field.
Methods: The coil is composed of a 16-channel self-decoupled loop transmit/receive array with a 112-loop receive-only (Rx) insert.
Purpose: To propose a novel method for parallel-transmission (pTx) spatial-spectral pulse design and demonstrate its utility for robust uniform water-selective excitation (water excitation) across the entire brain.
Theory And Methods: Our design problem is formulated as a magnitude-least-squares minimization with joint RF and k-space optimization under explicit specific-absorption-rate constraints. For improved robustness against off-resonance effects, the spectral component of the excitation target is prescribed to have a water passband and a fat stopband.
Article Synopsis
- The study aims to enhance ultrahigh-field brain imaging by evaluating the achievable signal-to-noise ratio (SNR) against the ultimate intrinsic SNR (uiSNR) at 10.5T, and exploring designs to improve SNR for better imaging results.
- A specialized 16-channel Tx/Rx array and a 64-channel receive-only array were created for use with the 10.5T MRI, with experiments confirming safe operational limits and comparisons of SNR at 10.5T and 7T.
- Results indicated that the technology can capture significant portions of uiSNR at 10.5T for high-resolution imaging, demonstrating its effectiveness for functional MRI, setting the stage for future advanced studies of the human
Article Synopsis
- The study aims to enhance brain imaging at 10.5 Tesla (T) by developing multichannel transmit and receive arrays to achieve the best possible signal-to-noise ratio (uiSNR).
- A new 16-channel transmit/receive array was created, alongside a 64-channel receive-only array, and underwent testing to ensure safety for human use, ultimately receiving FDA approval.
- Results showed that the new configuration significantly improved imaging quality, matching the effectiveness of lower-field setups while demonstrating high-resolution brain imaging capabilities for the first time at this field strength.