MRI Gradient Correction for Amplifier and Coil Deviations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) systems experience deviations from target magnetic field gradients due to factors like eddy currents, timing inaccuracies, and temperature fluctuations, leading to image artifacts and reduced image quality, especially in non-Cartesian recordings.
Innovation Solution
A method and apparatus using a machine-learning algorithm to create a trained function that corrects deviations in magnetic field gradients by compensating for amplifier and gradient coil non-linearities, employing input data such as target gradient information, amplifier signals, and patient-specific parameters to adapt the input signal and correct deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If gradient system is used to create magnetic field gradient, then spatially-dependent magnetic resonant frequency is achieved, but deviations from target gradient occur due to eddy currents, timing imprecisions, and amplification errors
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing correction values in lookup tables before actual imaging. The correction values are computed based on system characteristics and stored for rapid retrieval during gradient application, eliminating the need for real-time complex calculations and enabling timely compensation of gradient deviations
Solution Approach 2:
The patent implements feedback by measuring actual gradient deviations using field probes or dynamic field camera, comparing them against target gradients, and using the difference to adjust subsequent gradient applications. This closed-loop approach continuously refines gradient accuracy by feeding back correction information to the gradient system
2Measurement precision
If small field probes or dynamic field camera are used to measure emitted magnetic field gradients, then gradient deviations can be detected, but trajectory must be newly acquired for each reconstruction when parameters change
Solution Approach 1:
The patent pre-computes correction values and stores them in lookup tables before actual imaging sequences are executed. These correction values are derived from system characterization measurements and are ready for immediate application, eliminating the need for time-consuming real-time calculations during reconstruction
Solution Approach 2:
The patent creates a simplified model or representation of the gradient system behavior through lookup tables that capture the essential correction information. This copied representation allows rapid retrieval and application of correction values without requiring access to the full complexity of the original measurement data or repeated acquisition of trajectory information
3Productivity
If gradient system operates at high speed for non-Cartesian recordings, then imaging efficiency is improved, but temporal accuracy of magnetic field gradients deteriorates
Solution Approach 1:
The patent pre-calculates correction values for various gradient conditions and stores them in lookup tables, enabling rapid retrieval during high-speed imaging sequences. This eliminates the need for real-time computation during fast acquisitions, maintaining both speed and accuracy
Solution Approach 2:
The patent implements dynamic correction by selecting appropriate correction values from lookup tables based on actual gradient conditions during imaging. The system adapts the correction applied to each gradient pulse based on its specific parameters, enabling accurate compensation even during rapidly changing gradient sequences
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 provides precise and efficient correction of gradient deviations, improving image quality and reducing the need for time-consuming post-processing, suitable for clinical use.
Implementation Method 1
The MR system includes a gradient coil, which creates the magnetic field gradient with the aid of the amplifier output signal
Data Source
AI summary
A deviation from a target gradient of a magnetic field gradient created by an MR system is reduced or corrected. The MR system includes an amplifier, which amplifies an amplifier input signal and outputs an amplifier output signal, and a gradient coil, which creates the magnetic field gradient with the aid of the amplifier output signal. Input data is provided for a trained function trained by a machine-learning algorithm. The input data includes information about the target gradient of the MR system. Output data is created by the trained function with the aid of the input data. A gradient characterization function of the gradient coil is determined. The deviation from the target gradient of the magnetic field gradient created by the MR system is reduced and/or corrected. A deviation caused by the amplifier is reduced and/or corrected with the aid of the output data created, and a deviation caused by the gradient coil is reduced and/or corrected with the aid of the gradient characterization function.