Hybrid Iterative Image Reconstruction for CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current iterative reconstruction techniques for computed tomography (CT) imaging are computationally intensive, requiring significant time to converge to a solution, especially when dealing with heterogeneous data sets, and often necessitate reconstructing the entire field of view to achieve high image quality, which is inefficient for targeted reconstructions.
Innovation Solution
A hybrid optimization method that combines projection-based and voxel-based iterative reconstruction techniques, where a first reconstruction step using a projection-based technique quickly converges on low-frequency data, and a second step using a voxel-based technique focuses on high-frequency data, allowing for rapid convergence and improved image quality without the need for full field reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative reconstruction techniques are used to improve image quality, then image quality is improved, but computation time increases significantly
Solution Approach 1:
The patent segments the field of view into a region of interest (ROI) and surrounding regions. Iterative reconstruction is applied only to the ROI, while analytical methods are used for surrounding regions. This segmentation allows the system to achieve high image quality in the diagnostically important area without the computational burden of reconstructing the entire field of view iteratively, thus resolving the contradiction between image quality and computation time.
2Measurement precision
If full field of view reconstruction is performed to achieve high image quality, then image quality is improved, but reconstruction efficiency decreases
Solution Approach 1:
The patent extracts and focuses computational resources on the region of interest (ROI) by performing iterative reconstruction only in this localized area rather than reconstructing the entire field of view. This extraction of the critical diagnostic area from the full field allows high image quality to be achieved where needed while dramatically improving reconstruction efficiency by avoiding unnecessary computations in non-critical areas.
3Measurement precision
If iterative reconstruction is applied to heterogeneous data sets, then image quality is improved, but convergence time increases
Solution Approach 1:
The patent applies local quality by using different reconstruction methodologies tailored to different regions: iterative reconstruction with appropriate convergence criteria is applied to the heterogeneous ROI to ensure high image quality, while analytical methods are used for more homogeneous surrounding regions. This localized approach optimizes convergence time by avoiding unnecessary iterative processing in areas where it is less critical, while maintaining high image quality in the heterogeneous region of interest.
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
This approach significantly reduces computation time while maintaining high image quality by leveraging the strengths of different iterative methods, enabling faster reconstruction of images and accommodating various data types, including incomplete data sets, thus improving diagnostic efficiency.
Implementation Method 1
A computed tomography (CT) imaging system typically includes an x-ray source that projects fan- or cone-shaped x-ray beams through an object being imaged
Implementation Method 2
The beam is collimated to lie within an X-Y plane, or a set of such planes generally referred to as the 'imaging planes'
Implementation Method 3
Intensity of radiation from the beam received at the detector array depends on attenuation of the x-ray beam by the object
Data Source
AI summary
A method and system for image reconstruction of data acquired by a device such as computed tomography is provided. The method and system use a multi-stage statistical iterative reconstruction techniques to provide a three dimensional representation of the scanned object. In one embodiment, the first stage uses a projection-based reconstruction technique, such as Ordered Subset (OS) to converge on a solution for low frequency portion of the image. A subsequent stage uses a voxel-based reconstruction technique, such as Iterative Coordinate Descent (ICD), to converge on a solution for high frequency portions of the image. Systems and methods for reconstructing images from incomplete or partial projection data is also provided.