Magnetic Resonance Fingerprinting Signal Waveform Database Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional magnetic resonance fingerprinting methods require extensive and time-consuming signal comparisons with a large number of database signal waveforms, leading to complexity and increased computational burden, especially when determining multiple tissue parameters with fine discretization.
Innovation Solution
The method generates a reduced second signal waveform database from a first database, allowing selective comparison of magnetic resonance signal waveforms, optimizing the acquisition scheme for specific application cases by selecting a subset of database signal waveforms that exhibit strong differences, and using unsupervised learning to identify material classes and optimize scan parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of database signal waveforms are used for signal comparison to ensure accurate tissue parameter determination, then measurement precision is improved, but device complexity and computational burden increase
Solution Approach 1:
The patent segments the large database of signal waveforms into multiple smaller sub-databases or groups based on signal characteristics, tissue types, or parameter ranges. This segmentation allows the comparison process to be divided into multiple stages, where the acquired signal is first matched against a reduced subset of representative waveforms, thereby reducing computational complexity while preserving measurement precision through hierarchical or multi-stage comparison strategies.
Solution Approach 2:
The patent extracts and selects only the most relevant or representative signal waveforms from the complete database for inclusion in the comparison process. By identifying and removing redundant or less informative waveforms, the method reduces the number of comparisons needed while maintaining the ability to accurately determine tissue parameters, thus resolving the contradiction between precision and complexity.
2Measurement precision
If a large number of database signal waveforms are used for signal comparison to ensure accurate tissue parameter determination, then measurement precision is improved, but loss of time increases due to extensive computational processing
Solution Approach 1:
The patent performs preliminary actions by pre-processing the database signal waveforms before the actual comparison with acquired signals. This includes pre-calculating similarity metrics, organizing waveforms into hierarchical structures, or creating compressed representations that can be quickly compared. By preparing the database in advance, the method reduces the computational burden and time required during the actual tissue parameter determination process while maintaining measurement precision.
Solution Approach 2:
The patent implements a multi-stage comparison approach where the acquired signal waveform is first rapidly compared against a reduced subset or representative samples of the database using simplified criteria. Only after this initial screening does the method proceed to more detailed comparisons with a smaller number of candidate waveforms. This skipping strategy allows the system to quickly eliminate unlikely matches and focus computational resources on the most promising candidates, thereby reducing overall processing time while preserving accuracy.
3Measurement precision
If fine discretization is used for determining multiple tissue parameters to improve measurement precision, then measurement precision is improved, but device complexity and computational load increase
Solution Approach 1:
The patent addresses the complexity of fine discretization by introducing additional dimensions or levels to the comparison process. Instead of comparing against a single large database with fine discretization across all parameters simultaneously, the method organizes the database into hierarchical levels or dimensional groups (e.g., by tissue type, by parameter range, or by signal characteristics). This dimensional organization allows the system to navigate the parameter space more efficiently, reducing the apparent complexity while maintaining fine discretization precision through multi-dimensional indexing or hierarchical search strategies.
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 significantly reduces the complexity and computational load of signal comparisons, enabling faster and more efficient determination of tissue parameters while maintaining robust assignment and reducing scan time, resulting in a compact and effective magnetic resonance fingerprinting examination.
Implementation Method 1
Radio-frequency pulses, such as excitation pulses, are then radiated by suitable antennas of a radio-frequency antenna arrangement, which result in the nuclear spins of certain atoms having excited into resonance by the radio-frequency pulses
Implementation Method 2
During the relaxation of the nuclear spins, radio-frequency signals, referred to as magnetic resonance signals, are emitted, and are received by suitable radio-frequency antennas
Data Source
AI summary
In a method and magnetic resonance apparatus for implementing a magnetic resonance fingerprinting examination of an examination subject, a first signal waveform database containing multiple first database signal waveforms is provided to a processor, and a second signal waveform database is generated therefrom with a fewer number of signal waveforms therein than said multiple first database signal waveforms. A magnetic resonance signal waveform of a voxel of an examination region is acquired using a magnetic resonance fingerprinting method, and this is compared to the second database signal waveforms, and the comparison result is provided in electronic form.