Sinogram Shift Correction for PET Image Discontinuities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image quality issues arise when combining discrete sets of images acquired using a step-and-shoot imaging system with an extended axial field of view, particularly due to cradle misalignment or differential sag, leading to discontinuities that reduce clinical confidence in PET imaging systems.
Innovation Solution
A method and apparatus for retrospectively correcting image data by acquiring sinograms at different axial fields of view, determining object boundaries, measuring shifts between them, and generating corrected images using these shifts to align and reconstruct high-quality images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a step-and-shoot imaging system with extended axial FOV is used to acquire adjacent axial fields-of-view, then the productivity and coverage area are improved, but image discontinuities and quality deterioration occur due to cradle misalignment or differential sag
Solution Approach 1:
The patent applies preliminary action by measuring the shift between overlapping sinograms before final image reconstruction. The method identifies object boundaries in overlapping regions, calculates the misalignment shift, and applies correction to the sinogram data prior to reconstruction, preventing discontinuities in the final imageset
Solution Approach 2:
The patent replaces mechanical alignment precision requirements with a computational correction system. Instead of relying solely on precise mechanical cradle positioning, the system uses image processing algorithms to detect and correct for mechanical misalignments and sag in the overlapping sinogram regions
2Ease of operation
If cradle translation along an axis not aligned with the scanner axis or with differential sag occurs, then the ease of operation and adaptability are improved, but image discontinuities and quality are worsened
Solution Approach 1:
The patent implements feedback by using the measured shift from overlapping sinogram regions to correct the image reconstruction process. The system continuously monitors alignment by comparing overlapping fields-of-view and applies corrective transformations to maintain image continuity despite cradle misalignment or sag
Solution Approach 2:
The patent applies parameter changes by transforming the sinogram data using measured shift parameters. The method adjusts the positional parameters of the overlapping images based on detected misalignment, effectively compensating for cradle translation errors and differential sag through computational parameter adjustment
Data Source
AI summary
A method for retrospectively correcting data prior to image reconstruction using an imaging system, wherein the method includes acquiring a first sinogram of a first slice of an object at a first axial field of view and a second sinogram of the first slice of the object at a second axial field of view different than the first axial field of view, determining at least one boundary of the object in the first sinogram and the second sinogram at the first slice, measuring a shift between the first sinogram and the second sinogram using the determined boundaries, and generating a corrected image using the measured shift.