Binomial RF Pulse Sequence for MRI Water-Fat Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional Dixon-based MRI techniques face challenges in accurately separating water and fat signals due to issues like ghosting, background phase accrual, and sensitivity to eddy currents, which affect the robustness of field map estimation and image quality.

Innovation Solution

A magnetic resonance imaging system employing a binomial RF pulse sequence with independently phased RF flip angle components, followed by magnetic gradient spoiler pulses, to create a preparatory sequence for enhanced NMR species separation, allowing for more uniform signal acquisition across off-resonance NMR signals without altering the timing of the main MRI data acquisition sequence.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional Dixon-based techniques are used for water/fat separation, then the basic separation function is achieved, but ghosting and background phase accrual occur reducing image quality

Engineering Contradiction:
Improvewater/fat separation accuracyVSAvoidghosting and background phase
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary phase correction actions through the binomial RF pulse sequence before the main imaging acquisition. The first RF pulse creates transverse magnetization with a predetermined phase relationship between water and fat, and the second RF pulse nutates this magnetization back to longitudinal magnetization with a phase shift that compensates for background phase accrual, thereby preventing ghosting artifacts before they occur

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the phase parameters of the RF pulses in the binomial sequence to achieve robust separation. By setting specific phase differences between the first and second RF pulses and adjusting the phase evolution time, the system creates a preparatory sequence that is insensitive to time-dependent phase variations, thereby eliminating ghosting while maintaining separation accuracy

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If field map estimation is performed to improve water/fat separation, then separation accuracy improves, but the system becomes sensitive to eddy currents and phase instability

Engineering Contradiction:
Improvefield map estimation accuracyVSAvoidsensitivity to eddy currents
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent extracts and eliminates the need for field map estimation by using the binomial RF pulse sequence to directly create phase-differenced longitudinal magnetization. This approach removes the problematic field map calculation step that is sensitive to eddy currents, while still achieving accurate water/fat separation through the inherent phase relationships created by the preparatory pulses

Inventive Principle:
Principle #2Taking out (Extraction)

3Object-affected harmful factors

If phase cycling is implemented to reduce ghosting, then image quality improves, but the imaging time increases

Engineering Contradiction:
Improveghosting reductionVSAvoidimaging time
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent performs the phase correction action in advance through the binomial RF pulse sequence, eliminating the need for multiple phase-cycled acquisitions. The preparatory sequence establishes the correct phase relationships before data acquisition, allowing single-shot or efficient multi-echo imaging without the time penalty of acquiring and combining multiple phase-cycled datasets

Inventive Principle:
Principle #10Preliminary action

4Productivity

If aggressive readout methods like short echo-space FSE are used, then productivity increases, but phase instability and ghosting worsen

Engineering Contradiction:
Improveimaging speedVSAvoidphase instability
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent prepares the magnetization with the binomial RF pulse sequence to be inherently robust against phase instability before the aggressive readout begins. The phase-differenced longitudinal magnetization created by the preparatory sequence maintains stability even during rapid gradient switching in short echo-space FSE, enabling high productivity without sacrificing image quality

Inventive Principle:
Principle #10Preliminary action

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 enables improved fat/water separation images with reduced signal loss and increased robustness against time-dependent phase sources, facilitating the generation of high-quality images without the need for complex phase unwrapping or background phase removal, and can be used with aggressive readout methods like short echo-space FSE.

Implementation Method 1

using the phenomenon that resonant frequencies are slightly different between water and fat

Methodology Applied
Scientific EffectChemical shift:

Implementation Method 2

magnetic resonance imaging (MRI) processes and apparatus... separation of nuclear magnetic resonance (NMR) signals emanating from different NMR species

Methodology Applied
Scientific EffectNuclear magnetic resonance:

Data Source

PatentEP2508910B1Magnetic resonance imaging system and process
Publication Date: 2020.08.19 TOSHIBA MEDICAL SYST CORP
  • EP2508910B1 patent drawingFigure 1
  • EP2508910B1 patent drawingFigure 2
  • EP2508910B1 patent drawingFigure 3~4

AI summary

A magnetic resonance imaging system includes a sequence executing unit (30) and an image generating unit (42). The sequence executing unit (30) executes a prep applying portion including a first prepulse for exciting magnetization vectors of a plurality of NMR species having different resonant frequencies as transverse magnetizations and a second prepulse for nutating back the transverse magnetizations into longitudinal magnetization, the second prepulse being applied after a phase evolution time by which a predetermined phase difference is evolved between a subject NMR species and the other NMR species of the plurality of NMR species after the first prepulse, and execute a data acquisition sequence after the prep applying portion. The image generating unit (42) generates a separation image of a subject NMR species based on the data acquired by the data acquisition sequence.