AI Motion Estimation in PET Imaging Using Histo-Image Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current nuclear imaging systems face challenges in effectively correcting for both cyclic and non-cyclic patient motions during image acquisition, leading to noise and reduced image quality, as existing methods are specific to either cyclic or non-cyclic motions and cannot handle combined motion types effectively.
Innovation Solution
A novel method using artificial intelligence (AI) neural networks to process short frames of data from nuclear imaging systems, identifying a motion-free reference frame to estimate and correct for motion in other frames, applicable to both cyclic and non-cyclic motions, thereby improving signal-to-noise ratio and image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If PET data is separated by breathing phases to correct cyclic motion, then motion artifact is suppressed, but image quality suffers from greater noise due to reconstruction from less data
Solution Approach 1:
The patent segments the PET data into multiple short frames corresponding to different breathing phases, allowing separate reconstruction and motion estimation for each phase while maintaining the ability to combine results
Solution Approach 2:
The patent introduces an intermediary motion estimation step that operates on the segmented phase data before final image reconstruction, using the segmented frames as intermediaries to derive motion parameters that are then applied to improve the final image quality
2Reliability
If individual frame reconstruction and registration is performed to achieve motion artifact suppression, then signal-to-noise ratio is improved, but computational time and complexity increase
Solution Approach 1:
The patent divides the reconstruction process into segments where motion estimation is performed on segmented short frames, allowing parallel processing and reducing the computational burden on any single reconstruction operation
Solution Approach 2:
The patent performs preliminary motion estimation on segmented frames before the final image reconstruction, preparing motion parameters in advance that can be efficiently applied during the main reconstruction process, avoiding the need for repeated registration operations
3Adaptability or versatility
If separate motion correction methods are used for cyclic and non-cyclic motions, then each motion type can be corrected, but device complexity and processing steps increase
Solution Approach 1:
The patent implements a universal motion correction framework based on segmented frame analysis that can handle both cyclic and non-cyclic motions through the same fundamental process, eliminating the need for separate specialized methods for different motion types
Data Source
AI summary
A computer-implemented method for generating a motion corrected image is provided. The method includes receiving listmode data collected by an imaging system; producing two or more histo-image frames or two or more histo-projection frames based on the listmode data; providing the two or more histo-image frames or two or more histo-projection frames to an Artificial Intelligence (AI) system; receiving two or more AI reconstructed images from the AI system based on the two or more histo-image frames or the two or more histo-projection frames; and generating a motion estimation in reconstructed images by using a motion free AI reconstructed image frame as a reference frame.