MRI Apparatus Undersampled Signal Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic Resonance Imaging (MRI) techniques face challenges in generating accurate MR images in a short time, particularly when excessive undersampling is performed, leading to reduced image quality.
Innovation Solution
An MRI apparatus and method that store multiple MR signal data sets generated with varying MR parameters, allowing for undersampling and subsequent interpolation using matching parameter values to create accurate MR images, even with excessive undersampling, by extracting and adjusting data to minimize magnetic field inhomogeneity differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If undersampling is performed to reduce imaging time, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The system pre-calculates and stores multiple MR signal data sets with different parameter combinations before actual imaging. This preliminary preparation enables rapid selection and interpolation during undersampled imaging, allowing fast acquisition while maintaining accuracy through pre-computed reference data
Solution Approach 2:
The patent introduces an intermediate MR signal data set that serves as a bridge between the undersampled data and the final image. This intermediate data set is generated by selecting and combining stored signal data sets, then interpolated to reconstruct missing information, acting as a mediator that preserves image quality despite rapid acquisition
2Productivity
If excessive undersampling is performed to increase productivity, then imaging time is reduced, but manufacturing precision deteriorates
Solution Approach 1:
The system varies multiple MR parameters (echo time, repetition time, inversion time, flip angle) across different stored signal data sets. By changing these parameters systematically, the system creates diverse reference data that enables accurate interpolation even when severe undersampling is applied, maintaining image quality through parameter diversity
Solution Approach 2:
The patent extends the data space by incorporating multiple MR parameters as additional dimensions. Instead of relying solely on spatial sampling, the system utilizes parameter space diversity to reconstruct undersampled data, effectively adding temporal and parametric dimensions to compensate for spatial undersampling
3Measurement precision
If multiple MR signal data sets are stored with varying parameters to improve image accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system segments the MR signal data into multiple separate data sets, each corresponding to specific parameter combinations. This segmentation allows organized storage and efficient retrieval of relevant data subsets during interpolation, managing complexity through structured division rather than handling all data uniformly
Solution Approach 2:
The stored MR signal data sets serve multiple functions: they act as reference data for interpolation, provide parameter variation for reconstruction, and enable flexible selection based on specific imaging needs. This multi-functionality justifies the storage overhead by maximizing the utility of stored data across different undersampling scenarios
4Measurement precision
If interpolation is performed using extracted parameter values to maintain image accuracy, then measurement precision is improved, but loss of time increases
Solution Approach 1:
Parameter values are extracted and stored in advance along with the MR signal data sets. This preliminary extraction eliminates the need for time-consuming parameter calculation during the interpolation process, allowing rapid retrieval and immediate use of pre-computed parameter values
Solution Approach 2:
The system uses feedback from the undersampled data to select appropriate stored data sets and parameter combinations. By comparing the undersampled characteristics with stored reference data, the system efficiently identifies and retrieves the most suitable pre-computed data for interpolation, minimizing search and processing time
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 the generation of MR images similar to those from fully sampled data, thereby increasing image accuracy and efficiency.
Implementation Method 1
Magnetic resonance imaging (MRI) creates images by using information determined through the resonance of atomic nuclei exposed to a magnetic field. The resonance of atomic nuclei is a phenomenon where an atomic nucleus in a low energy state absorbs RF energy and is excited to a higher energy state when a specific radio frequency (RF) is incident on the atomic nucleus magnetized by an external magnetic field.
Data Source
AI summary
Provided is a magnetic resonance imaging (MRI) apparatus. The MRI apparatus includes: a storage configured to store a plurality of MR signal data sets generated by applying a plurality of values of a first MR parameter and a plurality of values of a second MR parameter to an MR signal data generation model; a data acquisition unit configured to acquire an MR signal data set for a specific position of an object by undersampling an MR signal, based on the values of the first MR parameter; and an image processor configured to extract an MR signal data set that matches the MR signal data set acquired by undersampling the MR signal (hereinafter referred to as the ‘undersampled MR signal data set’) from among the stored MR signal data sets, obtain a value of the second MR parameter for the undersampled MR signal data set based on the extracted MR signal data set, and interpolate unsampled MR signal data in the undersampled MR signal data set (hereinafter, referred to as the ‘interpolated MR signal data set’) by using the value of the second MR parameter.