PET Attenuation Map Correction via Movement Status Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance PET devices face challenges in recording high-quality PET image data sets for large areas due to movement artifacts caused by breathing and heart movements, as existing methods struggle to accurately correct for these movements during prolonged recording times, leading to diagnostic issues with small lesions.
Innovation Solution
A method that simultaneously records magnetic resonance data and PET raw data with movement status information, allowing for the creation of movement-resolved attenuation maps, which are applied to reconstruct high-quality PET image data sets by subdividing data into movement status classes using a pilot tone navigator and Dixon technique, enabling accurate movement correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If continuous patient couch movement is used to record large areas, then productivity is improved, but measurement precision deteriorates due to movement artifacts
Solution Approach 1:
The patent divides the continuous recording process into discrete movement status classes (e.g., breath-hold phases, different breathing states). PET raw data and magnetic resonance data are segmented according to these movement status classes, allowing separate processing and correction for each class. This segmentation enables the system to handle large recording areas continuously while maintaining precision by correcting movement artifacts for each segmented portion.
2Measurement precision
If patient couch moves slowly to record sufficient PET raw data, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary determination of movement status classes and assigns corresponding attenuation maps to different movement status classes before final PET image reconstruction. Magnetic resonance data is acquired and processed in advance to create movement-resolved attenuation maps, which are then applied during PET data reconstruction. This preliminary preparation eliminates the need for extremely slow couch movement, as the movement correction is handled through pre-computed attenuation maps rather than requiring prolonged data acquisition at very low speeds.
3Device complexity
If a single attenuation map is used for the entire recording area, then device complexity is reduced, but measurement precision deteriorates due to movement artifacts
Solution Approach 1:
The patent implements dynamic attenuation maps that are assigned based on movement status classes rather than using a single static attenuation map for the entire recording area. The system determines movement status from magnetic resonance data and dynamically selects or generates appropriate attenuation maps for each movement status class. This dynamic approach maintains relatively simple device architecture while significantly improving measurement precision by adapting the attenuation correction to the actual movement state during data acquisition.
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 results in higher quality PET image data sets with improved movement correction, enabling rapid whole-body screening and reducing artifacts, thus enhancing diagnostic accuracy.
Implementation Method 1
a material decomposition in particular can be carried out in magnetic resonance data of the anatomy of the patient, wherein specific attenuation values for the attenuation maps are assigned to each material
Implementation Method 2
using a pilot tone navigator and Dixon technique, enabling accurate movement correction
Data Source
AI summary
In a method for recording a PET image data set, an overall recording area is moved continuously through the FOV at a constant movement speed, an attenuation map of the overall recording area being used to reconstruct the PET image data record from the PET raw data. The magnetic resonance data of a slice of the patient currently located within the FOV and movement status information relating to a cyclical movement of the patient are recorded simultaneously with recording the PET raw data. A movement status class is assigned to the PET raw data and the magnetic resonance data in each case. Using the magnetic resonance data assigned to the different movement status classes, attenuation maps of the patient are determined for the different movement status classes and applied to the PET raw data assigned to the corresponding movement status class to reconstruct the PET image data set.