Scatter Sinogram Scaling for PET Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current nuclear medicine scanners, particularly PET scanners, face inaccuracies in scatter estimation due to the dominance of multiple scattering events in larger patients, leading to errors in image reconstruction, especially when the 'tail' portion of the sinogram used for scatter estimation becomes truncated or small, and existing methods like full Monte Carlo simulations are computationally intensive.
Innovation Solution
An image processing apparatus and method utilizing a combination of single-scatter simulation (SSS) and short Monte Carlo simulations to generate a scaled scatter sinogram, which approximates the total scatter contribution, eliminating the need for tail fitting and reducing computational time by deriving a scatter fraction ratio from primary and scattered photon pairs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full Monte Carlo simulation is used to estimate scatter contribution, then measurement precision is improved, but productivity deteriorates due to significant computational time requirements
Solution Approach 1:
The patent segments the scatter estimation process into two distinct parts: (1) a fast single-scatter simulation (SSS) that generates the basic scatter sinogram with correct spatial distribution, and (2) a scaling factor calculation that adjusts the overall magnitude to account for multiple scattering effects. This segmentation allows each component to be optimized independently, achieving both accuracy and speed.
Solution Approach 2:
Instead of performing a complete Monte Carlo simulation that models all scattering events, the patent uses a partial simulation approach that only models single scattering events (SSS). This partial action is then compensated by calculating a scaling factor that accounts for the missing multiple scattering contributions, achieving acceptable accuracy with significantly reduced computational effort.
2Productivity
If single-scatter simulation with tail fitting is used for scatter correction, then productivity is improved through faster processing, but measurement precision deteriorates when the tail portion is truncated or small in larger patients
Solution Approach 1:
The patent introduces a scaling factor as an intermediary parameter that bridges the gap between the simplified single-scatter simulation and the actual total scatter contribution. This scaling factor is calculated by comparing SSS results with a limited Monte Carlo simulation or measured data, and then applied to adjust the SSS scatter sinogram to better represent the true scatter distribution, even when tail fitting is not applicable.
3Productivity
If single-scatter simulation is used for scatter estimation, then productivity is improved through faster computation, but measurement precision deteriorates when multiple scattering contributes significantly to scattered events in larger patients
Solution Approach 1:
The patent changes the parameter being simulated from the complete scatter distribution (requiring full Monte Carlo) to the single-scatter component only (using SSS). This parameter change allows the use of faster computation methods while the missing multiple scattering information is recovered through the scaling factor adjustment, which scales the SSS results to match the actual scatter magnitude.
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 provides more accurate and quicker scatter estimation, improving image quality and reconstruction speed, while maintaining computational efficiency for all patient sizes without significant increases in processing time.
Implementation Method 1
a scatter simulation processor configured to use single-scatter simulation, SSS, for processing measured sinograms
Implementation Method 2
a fast Monte Carlo (MC) simulation that considers accurate photon migration and interactions due to photoelectric absorption and Compton scattering
Implementation Method 3
accurate photon migration and interactions due to photoelectric absorption and Compton scattering
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An image processing apparatus includes a scatter simulation processor which processes measured sinograms generated from imaging data acquired for an imaging subject by an imaging apparatus to produce a scatter sinogram that represents a shape of scatter contribution. A scatter scaling processor utilizesa Monte Carlo simulation to determine a scatter fraction and scales the scatter sinogram to generate a scaled scatter sinogram that matches the scatter contribution in the measured sinogram. A reconstruction processor reconstructs the imaging data into an image representation using the scaled scatter sinogram for scatter correction.