Deformed Attenuation Image Reconstruction for PET Motion Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tomographic imaging, particularly PET imaging, faces challenges in accurately correcting for motion-induced attenuation mismatch between CT and PET images, leading to artifacts and reduced quantitative integrity due to respiratory motion.
Innovation Solution
A method that involves generating anatomical images, processing them to obtain an initial attenuation image, and then reconstructing a corrected attenuation image by modeling the deformation of the initial image, which is used to calculate a final reconstructed emission image, thereby correcting for motion-induced attenuation mismatch.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Shape
If respiratory-gated acquisition is used to reduce motion blur, then image sharpness is improved, but image noise increases due to fewer counts per gate
Solution Approach 1:
The respiratory-gated acquisition divides the respiratory cycle into multiple gates, segmenting the PET data into time-resolved subsets. Each gate contains fewer counts (increasing noise) but the segmentation enables temporal sorting of data to reduce motion blur. The system processes each gate independently through deformation modeling to achieve both sharpness and acceptable noise levels.
2Measurement precision
If CT scan is performed with breath-hold to obtain attenuation image, then attenuation correction accuracy is improved, but respiratory motion artifacts are introduced in PET image reconstruction
Solution Approach 1:
The system transitions from a static attenuation correction model (breath-hold CT) to a dynamic model that accounts for respiratory motion during PET acquisition. The deformation field parameters are updated iteratively to match the dynamic position of anatomical structures between CT and PET images, correcting for respiratory motion artifacts while maintaining attenuation correction accuracy.
Solution Approach 2:
The system uses feedback from respiratory gating information to iteratively update the deformation field parameters. The respiratory gate timing provides feedback about the patient's respiratory phase, which is used to adjust the attenuation correction model dynamically, eliminating mismatches between the static CT-based attenuation map and the dynamic PET acquisition.
3Productivity
If conventional PET imaging is used without motion correction, then imaging speed is maintained, but quantitative integrity is degraded due to motion-induced attenuation mismatch
Solution Approach 1:
The system performs preliminary actions by acquiring respiratory-gated PET data and pre-processing the attenuation correction images through deformation modeling. The deformation field is estimated and applied in advance to the attenuation correction, preparing the data for accurate quantitative reconstruction without requiring additional real-time computational resources during the actual imaging process.
Data Source
AI summary
According to some embodiments, an emission tomography scanner may acquire emission scan data. One or more anatomical images may be generated using an anatomical imaging system, and the anatomical images may be processed to obtain an initial attenuation image. An emission image and a corrected attenuation image may be jointly reconstructed from the acquired emission scan data, the corrected attenuation image representing a deformation of the initial attenuation image. A final reconstructed emission image may then be calculated based on the reconstructed emission image and/or the corrected attenuation image. The final reconstructed emission image may then be stored in a data storage system and/or displayed on a display system.