Error Estimation in Quantitative Functional Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Quantitative functional imaging techniques, such as PET and SPECT, face challenges in accurately estimating errors in activity concentration measurements due to instrument limitations and inherent uncertainties, which affect the reliability of diagnostic and treatment planning decisions.
Innovation Solution
A method and system for estimating absolute error in quantitative functional imaging by varying variables in the system matrix and recalculating reconstructions to determine systematic and statistical errors, providing localized error information that can be used to improve diagnostic confidence and treatment planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If quantitative functional imaging is performed using PET or SPECT, then activity concentration can be estimated, but measurement error and uncertainty increase due to instrument limitations
Solution Approach 1:
The patent implements feedback by calculating error estimates from the reconstructed image data itself and using these error maps to guide subsequent imaging acquisitions and diagnostic decisions. The error information is fed back into the imaging process to identify regions needing higher precision measurement.
Solution Approach 2:
The patent changes parameters by varying system matrix variables (such as attenuation coefficients, scatter fractions, and resolution modeling parameters) to calculate systematic error estimates. This allows quantification of uncertainty without requiring additional physical measurements.
2Measurement precision
If error estimation is performed by altering system matrix variables and repeating reconstruction, then systematic error information is obtained, but computational time and processing complexity increase
Solution Approach 1:
The patent segments the error estimation process into separate systematic and statistical error components. Systematic error is estimated by altering specific system matrix variables, while statistical error is calculated from residual variations. This segmentation allows targeted computation rather than exhaustive reconstruction of all possible errors.
Solution Approach 2:
The patent applies partial action by altering only the most significant system matrix variables (such as attenuation and scatter parameters) rather than all possible parameters. This provides sufficient error estimation without the computational burden of exhaustively testing all parameter combinations.
3Measurement precision
If statistical error is calculated from image formation model, then measurement uncertainty is quantified, but the error estimate itself has associated error
Solution Approach 1:
The patent performs preliminary action by calculating error estimates during the image reconstruction process itself, before final diagnostic decisions are made. The error maps are generated as part of the standard reconstruction workflow, allowing error information to be available for immediate use in guiding acquisitions and interpretation.
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
The approach allows for the generation of error maps that indicate regions of high and low error, enhancing the reliability of diagnostic decisions and personalized dosimetry, thereby improving the accuracy of treatment assessments and planning.
Implementation Method 1
PET uses a detector ring about a patient to detect pairs of gamma rays emitted due to a collusion of a positron with an electron. The positron is generated by a radionuclide (tracer or radio-pharmaceutical) introduced into the patient.
Implementation Method 2
A gamma camera is rotated about the patient to detect gamma rays emitted by the radionuclide.
Data Source
AI summary
Systematic error is estimated as a function of location in quantitative functional imaging. The systematic error is estimated by perturbing the system matrix. The values of one or more variables of the system matrix are altered, and the reconstruction repeated. The differences in the resulting object or image data of the different reconstructions provide systematic error information as a function of location.