Medical Image Registration Using Confidence Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The quality of image registration for moving objects, such as the heart, is compromised due to image artifacts, leading to reduced motion estimation and reconstructed images with significant motion artifacts.
Innovation Solution
A registration apparatus and method that utilize a confidence map to determine the likelihood that image features are caused by the object's structure, rather than artifacts, by calculating confidence values based on weights and inverse variances of projection data values, and applying these to improve the registration transformation between images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If images are reconstructed without motion compensation for different times, then reconstruction speed is improved, but registration quality deteriorates due to image artifacts
Solution Approach 1:
The system performs preliminary motion estimation by registering images reconstructed without motion compensation, then uses this motion information to guide a second reconstruction pass with motion compensation. This preliminary action enables the system to quickly obtain initial registration results while preparing for subsequent refinement.
Solution Approach 2:
The patent introduces motion estimation as an intermediary step between initial image reconstruction and final motion-compensated reconstruction. The motion estimation acts as a mediator that translates initial registration results into motion parameters, which then guide the final high-quality reconstruction with reduced artifacts.
2Reliability
If motion compensation is applied in image reconstruction, then motion artifacts are reduced, but device complexity and computational load increase
Solution Approach 1:
The system applies motion compensation selectively rather than uniformly across all reconstruction steps. It performs motion estimation on initially reconstructed images and then applies motion compensation only where needed to correct specific artifacts, avoiding the excessive computational burden of full motion-compensated reconstruction from the outset.
Solution Approach 2:
The patent implements a dynamic, two-stage reconstruction process that adapts to the specific characteristics of the imaging data. The system dynamically determines motion parameters from initial reconstructions and adjusts the reconstruction process accordingly, rather than using a static, fixed computational approach.
3Loss of time
If registration is performed on images with motion artifacts, then processing time is reduced, but motion estimation quality deteriorates
Solution Approach 1:
The system performs preliminary registration on quickly reconstructed images to obtain initial motion estimates, then uses these estimates to guide a second, more accurate reconstruction and registration pass. This preliminary action enables time-efficient initial processing while setting the stage for higher quality refinement.
Solution Approach 2:
The patent implements a feedback loop where initial registration results inform motion estimation, which then guides subsequent reconstruction and registration. The system uses the output of the first registration pass as input for improving the second pass, creating a feedback mechanism that progressively enhances motion estimation quality.
Data Source
AI summary
The invention relates to a registration apparatus (14) for registering images comprising a unit (11) for providing a first and a second image of an object, such that an image element of the first image at a respective position has been reconstructed by multiplying projection data values of rays traversing the image element with weights and by backprojecting the weighted projection data values, a unit (12) for providing a confidence map comprising for different positions in the first image confidence values being indicative of a likelihood that an image feature is caused by a structure of the object, the confidence value being calculated as a sum of a function, which depends on the respective weight, over the rays traversing the respective image element, and a unit (13) for determining a transformation for registering the first and second image to each other under consideration of the confidence map.