MRI SAR Estimation Using Human Body Model Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for calculating specific absorption rate (SAR) in magnetic resonance imaging (MRI) do not accurately account for human body shape, leading to undervaluation of maximum local SAR and requiring excessive time for calculation.
Innovation Solution
A magnetic resonance imaging apparatus that selects a human body model corresponding to the subject and estimates local SARs at evaluation points based on the selected model and RF pulse parameters, ensuring these SARs fall below a local reference value, using processing circuitry to determine and adjust the RF pulse amplitude and phase.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SAR is calculated over the whole body of the subject for each setting of the input signal concerning the RF pulse, then the accuracy of SAR estimation is improved, but the calculation time increases
Solution Approach 1:
The patent divides the continuous SAR calculation process into discrete evaluation points based on human body shape characteristics. Instead of calculating SAR at every possible location, the method segments the body into representative evaluation points (such as head, torso, limbs) where maximum local SAR is likely to occur, thereby reducing calculation time while maintaining accuracy
Solution Approach 2:
The patent performs preliminary classification of composite electric field vectors into clusters based on degrees of similarity before SAR calculation. By pre-identifying representative evaluation points and clustering similar field patterns, the system prepares the data structure in advance to enable faster SAR estimation without sacrificing precision
2Productivity
If composite electric field vectors are classified into clusters and SARs are calculated at representative points, then the calculation time is shortened, but the maximum local SAR may be undervalued due to not considering human body shape
Solution Approach 1:
The patent applies local quality by selecting evaluation points that are specific to different body regions (head, torso, arms, legs) based on human body shape characteristics. Each body region has its own representative evaluation points where local SAR maxima are likely to occur, ensuring that the clustering method does not undervalue maximum local SAR in any specific anatomical region
Solution Approach 2:
The patent changes the parameters used for clustering from generic electric field vector similarity to similarity that incorporates human body shape characteristics. By modifying the clustering criteria to account for body shape, the system maintains accuracy in maximum local SAR estimation while still achieving computational efficiency through reduced calculation points
Data Source
AI summary
A magnetic resonance imaging apparatus according to the present embodiment includes processing circuitry and imaging control circuitry. The processing circuitry selects a human body model corresponding to a subject from human body models. The processing circuitry estimates local specific absorption rates (SARs) at evaluation points determined using the selected human body model, based on the selected human body model and an amplitude and/or phase of the RF pulse in an imaging protocol for magnetic resonance imaging scheduled to be performed on the subject. The processing circuitry determines whether or not the estimated local SARs fall below a local reference value. The imaging control circuitry executes the imaging protocol by using an amplitude and phase of the RF pulse which make the local SARs fall below the local reference value.