Emission Tomography Image Reconstruction Confidence Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Iterative image reconstruction in emission imaging, such as PET and SPECT, often results in incomplete convergence and noise amplification, leading to quantitative uncertainty in medical images, especially for small lesions and low radiopharmaceutical dosages, which can cause erroneous clinical conclusions.
Innovation Solution
A method that computes convergence quality metrics by generating intensity versus iteration curves and displaying convergence curves for regions of interest, allowing for improved confidence in image reconstruction and providing quantitative accuracy information, while reducing noise amplification and facilitating reduced iteration numbers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of iterations is increased to improve convergence, then quantitative accuracy is improved, but noise levels increase
Solution Approach 1:
The patent applies local quality by computing convergence metrics specifically for regions of interest (ROIs) rather than uniformly across the entire image. This allows the system to identify which specific regions have converged and which have not, enabling localized assessment of quantitative accuracy without being constrained by noise in non-ROI areas.
Solution Approach 2:
The patent implements partial action by performing a fixed number of iterations that may not achieve full convergence for all regions. Instead of iterating until complete convergence (which would amplify noise excessively), the system performs a predetermined number of iterations and then assesses which regions have sufficiently converged, accepting partial convergence as adequate for clinical purposes.
2Object-affected harmful factors
If the iterative reconstruction is stopped early to reduce noise, then noise levels are reduced, but convergence is incomplete leading to quantitative uncertainty
Solution Approach 1:
The patent implements feedback by computing convergence metrics after a fixed number of iterations and using this information to assess whether regions have converged sufficiently. The convergence metric provides feedback on the quality of reconstruction, allowing clinicians to understand the reliability of quantitative values in different regions without requiring further iterations.
Solution Approach 2:
The patent applies preliminary action by pre-defining regions of interest before the convergence assessment. This allows the system to focus computational resources on evaluating convergence specifically in clinically relevant areas, and to provide targeted convergence information to guide further imaging or treatment decisions.
3Productivity
If a fixed number of iterations is used for all regions, then reconstruction efficiency is improved, but small lesions may not be fully converged
Solution Approach 1:
The patent applies segmentation by dividing the image into regions of interest and non-ROI areas, and further segmenting ROIs into subregions based on convergence status. This allows the system to process and assess different regions independently, identifying which specific areas have converged and which require additional iterations or attention.
Solution Approach 2:
The patent implements parameter changes by computing convergence metrics that quantify the state of convergence for each region. These metrics provide information about whether the current number of iterations is sufficient for each specific region, allowing adaptive adjustment of iteration parameters for different anatomical or pathological regions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A non-transitory storage medium stores instructions readable and executable by an imaging workstation (14) including at least one electronic processor (16) operatively connected with a display device (20) to perform an image reconstruction method (100). The method includes: reconstructing imaging data acquired by an image acquisition device (12) using an iterative image reconstruction algorithm to generate at least one reconstructed image (22); delineating one or more contours (26) of the at least one reconstructed image to determine a region of interest (ROI) (24) of the at least one reconstructed image; computing at least one quality metric value (30) of the ROI, the at least one quality metric value including at least one of a convergence quality metric, a partial volume effect (PVE) quality metric, and a local count quality metric; and displaying, on the display device, the at least one quality metric value and the at least one reconstructed image showing the ROI.