4D Medical Image Motion Correction via Gate Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical imaging systems face challenges in maintaining image quality due to motion artifacts, particularly respiratory motion, which can lead to errors in lesion size, location, and quantification, as existing methods using reference gates at the end of the breathing cycle may not include optimal information, resulting in reduced pixel intensity and image quality.
Innovation Solution
A method and system for reducing motion-related artifacts by obtaining a 4D image dataset, sorting it into gates, quantifying features using metrics, selecting a reference gate based on these metrics, and registering other gates to the reference gate using rigid or non-rigid transformations to generate an improved image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the reference gate is selected at the end of the patient's breathing cycle, then the respiratory motion is captured, but the image quality is reduced due to fewer pixels and reduced pixel intensity
Solution Approach 1:
The patent changes the selection criterion for the reference gate from a fixed temporal position (end of breathing cycle) to an optimization-based selection using image quality metrics. The reference gate is selected by evaluating metrics such as pixel intensity, number of pixels, and image sharpness across multiple gates, thereby changing the parameter basis from temporal to quality-based selection.
Solution Approach 2:
The patent introduces a feedback mechanism where image quality metrics are calculated for each gate, and the reference gate selection is adjusted based on these metric evaluations. The system uses the calculated metrics to iteratively identify the optimal reference gate that maximizes image quality while maintaining respiratory motion representation.
2Ease of operation
If gates are registered to a reference gate with reduced pixel information, then registration is performed, but the resultant image quality is degraded
Solution Approach 1:
The patent performs preliminary evaluation of image quality metrics for all gates before selecting the reference gate. By pre-calculating metrics such as pixel intensity and image sharpness for each gate, the system ensures that the reference gate is optimally selected before the registration process begins, preventing quality degradation from the outset.
Solution Approach 2:
The patent replaces the conventional mechanical/temporal-based reference gate selection with an information-based selection system. Instead of relying on the physical timing of the breathing cycle, the system substitutes a metric-driven selection process that evaluates the information content and quality of each gate, thereby substituting mechanical timing with informational optimization.
Data Source
AI summary
A method for reducing, in an image, motion related imaging artifacts includes obtaining a four-dimensional (4D) image dataset of a region of interest, sorting the 4D image dataset into a plurality of gates, generating an image of the region of interest using the plurality of gates, quantifying a selected feature using at least one metric, selecting a reference gate from the plurality of gates based on the quantification of the feature, and registering one or more of the plurality of gates to the reference gate to generate an image of the selected feature. A system and non-transitory computer readable medium configure to implement the method are also described herein.