Magnetic Resonance Fingerprinting B1+ Encoding via RF Phase Offsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional magnetic resonance fingerprinting methods face challenges in accurately determining T1, T2, and B1+ relaxation parameters due to RF transmission field inhomogeneities and the need for additional B1+ maps, which increase scan time and introduce inaccuracies.
Innovation Solution
Incorporating B1+-dependent information into the MR signal curve by using RF pulse sequences with phase offsets, allowing the local amplitude of the B1+ field to be translated into phase variations of the MR signals, enabling unambiguous association of T1, T2, and B1+ parameters without the need for separate B1+ maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate B1+ map acquisition is performed to correct RF transmission field inhomogeneities, then measurement precision of relaxation parameters is improved, but loss of time increases due to extended scan duration
Solution Approach 1:
The patent combines the B1+ mapping function and the MR fingerprinting measurement into a single integrated pulse sequence. The RF pulse sequence with phase offsets simultaneously encodes both the B1+ field information and the tissue relaxation signals, eliminating the need for separate B1+ map acquisition and reducing total scan time while maintaining precision.
Solution Approach 2:
The RF pulse sequence is designed to serve multiple functions: it acts as both the excitation sequence for MR fingerprinting and the encoding sequence for B1+ mapping. By making the pulse sequence universal, the patent eliminates redundant measurements and achieves both B1+ correction and relaxation parameter measurement in a single scan.
2Measurement precision
If multiple two-dimensional dictionaries are used for various B1+ values, then measurement precision of B1+ and relaxation parameters is improved, but device complexity increases due to expanded dictionary structure
Solution Approach 1:
Instead of creating multiple separate two-dimensional dictionaries for different B1+ values, the patent introduces a phase dimension to the signal model. By encoding B1+ information in the phase of the MR signal through RF pulse phase offsets, the patent transforms a multi-dimensional dictionary problem into a single dictionary with phase-sensitive signal matching, significantly reducing complexity.
Solution Approach 2:
The patent changes the parameter encoding strategy from using separate dictionaries for different B1+ values to using phase modulation within a single dictionary. The RF pulse phase offsets cause the B1+ field variations to manifest as phase changes in the signal, allowing a single dictionary to handle all B1+ conditions through phase-sensitive matching.
3Reliability
If B1+ map acquisition is performed upstream or downstream of MRF measurement, then reliability of B1+ correction is improved, but loss of time increases and susceptibility to subject movement errors worsens
Solution Approach 1:
The patent merges the B1+ mapping and MRF measurement into a single simultaneous process. The RF pulse sequence with phase offsets is applied during the MRF measurement itself, encoding B1+ information directly into the fingerprinting signals. This eliminates the temporal separation between B1+ mapping and MRF, preventing subject movement artifacts and reducing total time.
Solution Approach 2:
The patent maintains continuous encoding of both B1+ and relaxation information throughout the entire MRF measurement duration. By continuously applying the phase-offset RF pulses during the pseudorandomized flip angle series, the system continuously accumulates both B1+ and tissue relaxation data without interruption, ensuring reliability while minimizing scan time.
4Loss of time
If RF pulse sequence with phase offsets is used to encode B1+ information, then loss of time is reduced by eliminating separate B1+ mapping, but device complexity increases due to modified pulse sequence
Solution Approach 1:
The patent modifies the RF pulse sequence by introducing phase offsets between successive pulses. This parameter change (adding phase modulation) enables B1+ encoding without requiring fundamentally new hardware or complex multi-dimensional dictionaries. The phase offset is a simple controllable parameter that can be implemented in standard MRI systems, achieving time reduction with minimal increase in actual device complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows simultaneous measurement of all three parameters within the same scan time, reducing errors from inaccurate B1+ maps and enhancing robustness to subject movement, while integrating B1+ information into the MR fingerprinting process.
Implementation Method 1
translate variations of the local amplitude of the B1+ field into a corresponding variation of the phase of the acquired MR signals by the use of special RF excitation pulses
Implementation Method 2
nuclear spins are excited with an RF pulse sequence in order to give the received (detected) MR signal a phase that is B1+-dependent
Data Source
AI summary
In a method and apparatus for magnetic resonance (MR) fingerprinting, an MR signal is acquired from a subject by radiating radio-frequency (RF) energy that produces an RF transmission field that has a localized amplitude in the subject. The RF energy is radiated with an RF pulse configuration that maps the localized RF field amplitude in the phase of the MR signal from the subject. The detected MR signal is compared to a source of stored MR signal physical or theoretical models that respectively map different localized RF transmission field information in the respective phase thereof, the stored models being respectively for different substances. A substance in the subject from which the detected MR signal curve originated is identified by comparing the detected MR signal curve to the stored models to identify a best match.