Trained Function for Faster B0 and B1 Field Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining magnetic field data, such as B1 and B0 fields, in magnetic resonance imaging are time-consuming and inaccurate.
Innovation Solution
A computer-implemented method using a trained function, specifically an optimized trained function (OTF), which includes training image data correction and iterative optimization to provide accurate magnetic field data, reducing spatial amplitude variations and inhomogeneities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard methods are used to determine magnetic field data, then measurement precision is maintained, but loss of time increases
Solution Approach 1:
The system performs preliminary actions by training a neural network model in advance using corrected training image data and reference magnetic field data. The trained model is then deployed to rapidly predict magnetic field data without requiring time-consuming iterative calculations during actual MRI measurements, thus resolving the contradiction between measurement precision and time loss.
Solution Approach 2:
The invention creates a computational copy of the magnetic field determination process through a trained neural network model. Instead of performing complex iterative calculations each time magnetic field data is needed, the system uses the trained model to generate accurate predictions quickly, maintaining measurement precision while dramatically reducing computation time.
2Reliability
If standard methods are used to determine magnetic field data, then reliability is maintained, but productivity decreases
Solution Approach 1:
The invention replaces the traditional mechanical/iterative calculation system with a neural network-based computational system. The trained model substitutes complex iterative algorithms, providing reliable magnetic field data predictions instantaneously, thereby maintaining reliability while significantly improving productivity and measurement throughput.
Solution Approach 2:
The system changes the computational parameters by transitioning from iterative numerical methods to a pre-trained neural network inference process. This parameter change enables the system to maintain reliable magnetic field data determination while dramatically reducing processing time and increasing overall measurement productivity.
3Measurement precision
If iterative optimization is performed to improve magnetic field data accuracy, then measurement precision improves, but loss of time increases
Solution Approach 1:
The system performs the iterative optimization process in advance during the model training phase using corrected training image data and reference magnetic field data. Once trained, the neural network model can rapidly predict magnetic field data without requiring further iterative optimization, thus achieving high measurement precision while eliminating time loss during actual measurements.
Data Source
AI summary
A method for providing magnetic field data includes receiving image data as input data of a trained function, and applying the trained function to the image data. The trained function is trained based on a data fidelity of image data corrected using the magnetic field data, and based on at least one assumption about at least one attribute of the magnetic field data. The method includes providing the magnetic field data as output data of the trained function.