MRI Parametric Map Generation Using Artificial Neural Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating T2 maps in MRI, such as multi-echo acquisition and Bloch equation-based models, suffer from poor prediction accuracy and long computation times due to the complexity of radio frequency pulse inhomogeneity and exponential decay model limitations.
Innovation Solution
An MRI apparatus and method utilizing an artificial neural network (ANN) for training with simulated MR signal data and RF slice profiles to calculate characteristic parameter values, allowing for faster and more accurate generation of parametric maps like T2 maps by correcting errors through a dictionary-based approach.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Bloch equation or EPG model is used for T2 quantification, then prediction accuracy is improved, but computation time increases significantly
Solution Approach 1:
The patent pre-calculates and stores RF slice profiles and signal modeling data in a lookup table before actual T2 mapping. During clinical operation, the system simply queries this pre-computed table rather than performing complex Bloch equation calculations, thus achieving high accuracy with minimal computation time.
Solution Approach 2:
The patent creates a simplified copy of the complex Bloch equation model by pre-computing signal responses for various T2 values and storing them in a lookup table. This copy allows rapid retrieval of accurate T2 values without repeating the full computational process.
2Reliability
If multi-echo acquisition method is used, then T2 map generation is enabled, but prediction accuracy deteriorates due to RF pulse inhomogeneity
Solution Approach 1:
The patent incorporates RF slice profiles as additional parameters in the signal modeling process. By accounting for RF pulse inhomogeneity through these profiles, the system corrects the exponential decay model's inaccuracies and improves T2 quantification precision while maintaining multi-echo acquisition capability.
3Measurement precision
If RF slice profiles are incorporated into signal modeling, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent pre-calculates RF slice profiles and stores them in lookup tables before actual imaging. This eliminates the need for complex real-time calculations during clinical operation, reducing the apparent system complexity while maintaining high measurement precision.
Data Source
AI summary
Imperfect RF pulses in a multi-spin-echo (MSE) sequence disturb prediction of relaxation times. Provided are a magnetic resonance imaging (MRI) apparatus and method of operating the same, whereby a characteristic parameter value may be acquired from MR signal data via training using an artificial neural network (ANN) and a parametric map may be generated based on the acquired characteristic parameter value. The ANN may be trained to compensate for imperfect RF pulses while providing reduced computation times to produce an output image.