CT Image Detail Recovery via Forward Projection Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital x-ray imaging systems face challenges in preserving image detail and quality due to oversmoothing caused by AI correction processes, especially when original projection images are not available for iterative reconstruction.
Innovation Solution
A digital x-ray imaging system that uses an initial volume as input to reconstruct forward projections and applies an AI model to correct the volume, followed by an iterative reconstruction process using these forward projections and the corrected volume to optimize image quality without relying on original projection images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If AI correction is applied to remove artifacts and improve image quality, then image quality is improved, but image detail is lost due to oversmoothing
Solution Approach 1:
The patent segments the image processing into distinct stages: initial reconstruction, AI correction, forward projection generation, and iterative reconstruction. This segmentation allows the AI to focus on artifact removal while the iterative reconstruction stage specifically targets detail preservation through data consistency enforcement.
Solution Approach 2:
The patent introduces forward projections as an intermediary element that bridges the AI-corrected volume and the iterative reconstruction process. These forward projections serve as a mediator that carries diagnostic information from the original data through the AI processing while maintaining consistency for detail recovery.
2Loss of information
If iterative reconstruction is applied to restore image detail, then image detail is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary AI-based artifact correction before the iterative reconstruction process. This preliminary action removes major artifacts upfront, allowing the subsequent iterative reconstruction to focus specifically on detail recovery with fewer iterations needed, thereby reducing overall processing time.
Solution Approach 2:
The patent maintains continuous useful action by generating forward projections from the AI-corrected volume and using them immediately in the iterative reconstruction process. This continuous workflow ensures that diagnostic information is preserved and utilized throughout the processing chain without interruption or redundant computations.
3Loss of information
If original projection images are stored for iterative reconstruction, then image detail can be restored, but data storage requirements and system complexity increase
Solution Approach 1:
The patent creates forward projections as copies derived from the AI-corrected volume, which then serve the purpose of original projections in the iterative reconstruction process. These synthesized forward projections contain the necessary diagnostic information without requiring storage or retrieval of the actual original projection images, thereby simplifying the system architecture.
Data Source
AI summary
An image correction system and method takes an initial volume reconstructed from a number of projections obtained by the imaging system as the sole input to the image correction system. In a first step, the image correction system reconstructs a number of reconstructed or forward projections from the initial volume. In a second and optionally concurrent step, the image correction system forms a corrected volume by applying an artificial intelligence/deep learning/convolutional neural network model to the initial volume. In a third step, the image correction system employs the forward projections and the corrected volume as inputs to an iterative reconstruction process to achieve an optimized volume as an output from the image correction system. The use of the initial volume as the only input to the image correction system simplifies the computational processes of the image processing system while providing an optimized image having improved detail and image quality.