PET Normalization Correction via Geometric Symmetry Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current PET system normalization correction methods require extensive data collection and are inaccurate when insufficient coincidence event data is available, leading to data distortion and pseudo-shadows in images due to sensitivity differences between detector channels.
Innovation Solution
A normalization correction method that calculates geometric symmetry values, groups lines of response based on symmetry, and iteratively calculates geometric and crystal efficiency factors to determine high-precision normalization factors using a maximum-likelihood estimation method, allowing for accurate normalization with reduced data requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional normalization correction methods are used to ensure accuracy, then measurement precision is improved, but loss of time increases due to extensive data collection requirements
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing geometric symmetry values for all possible lines of response based on the detector's geometric configuration. This pre-computed geometric information is then reused during normalization correction, eliminating the need to collect extensive empirical data for each correction calculation, thus reducing data collection time while maintaining accuracy
Solution Approach 2:
The patent substitutes the traditional mechanical/data-intensive empirical measurement system with a computational geometry-based system. By using geometric symmetry calculations and maximum-likelihood estimation algorithms, the method replaces extensive physical data collection with mathematical computations, significantly reducing the time required while preserving measurement precision
2Loss of time
If data collection is reduced to save time, then loss of time is improved, but measurement precision deteriorates due to insufficient coincidence event data
Solution Approach 1:
The patent introduces geometric symmetry values as an intermediary that bridges the gap between limited empirical data and accurate normalization correction. These pre-calculated geometric parameters serve as mediators that enable the maximum-likelihood estimation algorithm to achieve accurate results even with insufficient coincidence event data, thus maintaining measurement precision while reducing data collection requirements
Solution Approach 2:
The patent changes the approach from relying on large volumes of empirical coincidence event data to using pre-calculated geometric parameters and iterative mathematical estimation. By transforming the problem from data-intensive empirical measurement to computation-intensive geometric calculation with iterative refinement, the method achieves accurate normalization with minimal data while reducing calculation time through the use of efficient algorithms
3Device complexity
If sensitivity differences between detector channels are not corrected, then device complexity is reduced, but image uniformity deteriorates leading to data distortion and pseudo-shadows
Solution Approach 1:
The patent applies segmentation by dividing the normalization correction process into distinct components: geometric symmetry calculation, crystal efficiency factor determination, and iterative maximum-likelihood estimation. This segmented approach systematically addresses sensitivity differences between detector channels through structured correction factors, improving image uniformity while keeping the correction process organized and manageable
Solution Approach 2:
The patent creates a universal normalization correction method that can be applied to any PET detector configuration by using geometric symmetry principles that are independent of specific detector arrangements. The method universally handles sensitivity differences across all detector channels through a standardized iterative correction process, improving image uniformity without requiring complex channel-specific adjustments
Data Source
AI summary
Provided is a normalization correction method for a PET system a detector including crystals. The method includes: obtaining a coincidence event dataset including a plurality of lines of response; calculating a geometric symmetry value of each line of response; grouping the plurality of lines of response according to the geometric symmetry value, where a plurality of lines of response in a group have a same geometric symmetry value; calculating a number of coincidence events of a same group through accumulation; cyclically calculating, according to the calculated number of the coincidence events for the same group, a geometric factor value of each group, where each line of response in the same group of lines of response has a same geometric factor value; cyclically calculating crystal efficiency factor values of the crystals; and calculating normalization factor values of the lines of response respectively, according to the geometric factor value and the crystal efficiency factor values.