Sinogram Pseudo-Slice Insertion for PET Detector Gap Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern PET scanners with detector blocks separated by gaps face mis-registration errors during image stitching due to unaccounted gaps in image reconstruction, leading to incorrect positioning of objects in PET/CT or PET-PET images, and current methods to correct these errors compromise computational efficiency.
Innovation Solution
The method involves generating a sinogram based on photon coincidence events and selectively inserting pseudo-slices into the sinogram to account for gaps between detector blocks, allowing for reduced positional errors without modifying the image reconstruction algorithm or detector geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If gaps between detector blocks are eliminated, then manufacturing precision of detector geometry is improved, but device complexity and cost increase substantially
Solution Approach 1:
The patent introduces pseudo-slices as an intermediary computational element to represent detector gaps in the sinogram data. These pseudo-slices act as a mediator between the physical detector geometry with gaps and the image reconstruction process, allowing the reconstruction algorithm to account for gap-induced positioning errors without requiring physical modification of the detector system. This resolves the contradiction by maintaining the simple detector geometry while achieving precise image positioning through data-level intervention.
2Manufacturing precision
If full detector geometry is accounted for in reconstruction, then manufacturing precision of image positioning is improved, but productivity of image reconstruction deteriorates
Solution Approach 1:
The patent segments the sinogram data by inserting pseudo-slices at specific locations corresponding to detector gaps. This segmentation allows the reconstruction algorithm to process most data using efficient symmetric operations while only requiring additional correction at specific segmented regions. The pseudo-slices divide the problem into manageable parts: standard reconstruction for continuous regions and gap correction for discrete pseudo-slice locations, thereby maintaining high productivity while achieving precise positioning.
Solution Approach 2:
The patent applies gap correction locally at pseudo-slice locations rather than throughout the entire reconstruction process. By identifying specific positions where detector gaps occur and applying correction only at those locations through pseudo-slice insertion, the method maintains high reconstruction speed for the majority of data while achieving precise positioning correction where needed. This localized approach resolves the contradiction between precision and speed.
3Ease of manufacture
If detector blocks are separated by gaps, then ease of manufacture and assembly is improved, but measurement precision of image positioning deteriorates
Solution Approach 1:
The patent creates a computational copy of the gap structure through pseudo-slices in the sinogram data. Rather than attempting to physically eliminate gaps for the sake of positioning precision, the method copies the gap information into the data domain, where it can be explicitly accounted for during reconstruction. This computational copying allows the physical detector to remain easy to manufacture with gaps, while the image positioning precision is restored through data processing that mirrors the physical gap structure.
Data Source
AI summary
Methods and systems are provided for correcting positional errors in an image arising from gaps in a detector assembly. In one embodiment, a method comprises generating a sinogram based on a plurality of photon coincidence events, selectively inserting one or more pseudo-slices into the sinogram, and generating an image based on the sinogram including the one or more pseudo-slices. In this way, positional errors may be reduced without modifying an image reconstruction algorithm to include a full detector geometry or modifying the detector geometry itself.