Iterative CT Reconstruction with System Optics Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CT scanners face challenges in reconstructing clinically viable images due to issues like beam hardening, temporal resolution, noise, and poor detector response, especially in low-dose applications, where maintaining image fidelity and distinguishing small features is difficult.
Innovation Solution
The implementation of iterative reconstruction with system optics modeling using spatially and view-variant low-pass filters, which model the blur caused by the CT imaging system, reducing computational complexity by applying filters in the reprojection or image domain, and using a calibration phantom to measure the point spread function for improved image sharpness and noise reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional iterative reconstruction methods are used, then image reconstruction is achieved, but image sharpness deteriorates due to system optics blur
Solution Approach 1:
The patent introduces a system optics model as an intermediary component that characterizes the blur introduced by the CT imaging system. This model includes parameters such as source size, detector element size, and geometric relationships, which are used to compute a point spread function (PSF). The PSF acts as a mediator between the true image and the blurred measured data, enabling deconvolution during iterative reconstruction to restore sharpness without requiring overly complex computational methods.
Solution Approach 2:
The patent employs parameter changes by adjusting the point spread function characteristics based on system optics parameters. By varying the PSF parameters (such as spatial frequency cutoffs and blur kernels) during the iterative reconstruction process, the system optimizes the balance between noise reduction and sharpness preservation. This allows the reconstruction algorithm to adapt to different imaging conditions and maintain image quality without excessive computational burden.
2Reliability
If noise reduction filters are applied during iterative reconstruction, then noise is reduced, but image sharpness and edge preservation deteriorate
Solution Approach 1:
The patent applies local quality by implementing spatially variant filtering that adapts to different regions of the image. Instead of using a uniform filter, the system computes local point spread functions and applies corresponding deconvolution operations in different spatial regions. This allows aggressive noise reduction in homogeneous areas while preserving edges and fine structures in critical regions, thereby maintaining both noise performance and edge sharpness simultaneously.
Solution Approach 2:
The patent incorporates feedback mechanisms where the iterative reconstruction algorithm continuously evaluates the reconstructed image against the measured data and adjusts the deconvolution parameters accordingly. The system uses feedback from residual analysis and convergence monitoring to dynamically tune the balance between noise suppression and sharpness preservation, preventing over-smoothing while maintaining reliable noise reduction throughout the reconstruction process.
3Manufacturing precision
If system optics modeling is implemented, then image quality improves, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing the system optics model parameters and point spread function characteristics before the iterative reconstruction process begins. The system optics parameters (source size, detector geometry, magnification factors) are characterized in advance, and the resulting PSF templates are prepared and cached. This preliminary preparation eliminates the need to compute complex system optics corrections during each iteration, significantly reducing the per-iteration computational burden while maintaining high image quality.
Solution Approach 2:
The patent uses copying by creating simplified representations of the complex system optics behavior through the point spread function model. Instead of performing full physical optics simulations during reconstruction, the system copies the essential blur characteristics into a mathematical model (PSF) that can be efficiently applied through standard convolution operations. This copying approach captures the necessary system optics effects without the computational intensity of complete physical modeling.
Data Source
AI summary
A CT imaging apparatus has processing circuitry that is configured to obtain projection data collected by a CT detector during a scan of an object. The processing circuitry is also configured to perform iterative reconstruction of the projection data to generate a current image. The iterative reconstruction includes filtering forward-projected data during backprojection or filtering image data prior to forward projection to model system optics. The processing circuitry is also configured to combine the current image with a previously-obtained image to generate an updated image.