MR-Based Attenuation Correction in PET/MR Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current MR-based attenuation correction techniques for PET imaging suffer from long data acquisition times, motion artifacts, and signal cancellations due to restrictive echo time requirements, which limit the resolution and accuracy of attenuation maps.

Innovation Solution

The use of the mDixon technique with flexible echo times and the combination of UTE pulse sequencing allows for shorter repetition times, reduced data acquisition times, and avoidance of signal cancellations, enabling more accurate and efficient MR-based attenuation correction by optimizing echo times and phase differences between water and fat echoes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional Dixon technique is used with optimized echo times for water-fat separation, then spatial resolution is improved, but data acquisition time increases beyond breath hold capability

Engineering Contradiction:
Improvespatial resolutionVSAvoiddata acquisition time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent changes the echo time parameters from traditional Dixon optimization (focused on water-fat separation quality) to a new optimization criterion (focused on minimizing total acquisition time while maintaining sufficient separation). This allows the system to operate within breath-hold time limits while still achieving adequate tissue classification for attenuation correction.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If long repetition times are used in Dixon technique to achieve complete water-fat separation, then tissue classification accuracy is improved, but respiratory motion artifacts increase

Engineering Contradiction:
Improvetissue classification accuracyVSAvoidrespiratory motion artifacts
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent performs the water-fat separation and tissue classification before the subject resumes normal breathing, completing the critical measurement phase during the breath-hold period. This preliminary action prevents respiratory motion from degrading the quality of the attenuation correction map.

Inventive Principle:
Principle #10Preliminary action

3Illumination intensity

If conventional echo time selection is used to maximize water-fat signal separation, then signal contrast is improved, but signal cancellations occur in certain tissue regions

Engineering Contradiction:
Improvesignal contrastVSAvoidsignal stability
Core Design Contradiction:
Illumination intensityVSReliability

Solution Approach 1:

The patent uses a first echo time that provides sufficient (but not maximal) water-fat signal separation, accepting that some signal contrast is sacrificed to avoid the complete signal cancellation that occurs at specific echo times. The second, shorter echo time provides complementary information that helps resolve ambiguities and improves overall tissue classification reliability.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3049818B1Mr-based attenuation correction in pet/mr imaging with dixon pulse sequence
Publication Date: 2021.08.11 KONINKLIJKE PHILIPS NV
  • EP3049818B1 patent drawingFigure 1
  • EP3049818B1 patent drawingFigure 2

AI summary

A medical imaging system (10) includes a nuclear imaging system (62), a timing optimization unit (40), a magnetic resonance (MR) scanner (12), an MR reconstruction unit (38), and an attenuation map unit (50). The nuclear imaging system (62) receives nuclear decay data and generates at least one nuclear image (64) of a first resolution based on the received nuclear decay data of an imaged subject (16) and an attenuation map (52). The timing optimization unit (40) which selects a first and a second echo time for a modified Dixon (mDixon) pulse sequence and a sufficient number of repetition times (TRs) to generate an image of the subject (16) of at least a first resolution, with the phase angle difference between water and fat at the first and the second echo time being unequal to 0° and 180°. The MR scanner (12) applies the sequence to the subject (16) and receives MR data (32) from the subject. The MR reconstruction unit (38) reconstructs at least one MR image (44) based on the MR data (32). The attenuation map unit (50) constructs the attenuation map (52) based on the reconstructed MR image (44).