CT Reconstruction Combining Full-View and Detailed-View Scans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CT reconstruction methods suffer from poor image quality and low efficiency due to high data requirements and incomplete data causing artifacts, particularly in imaging fine structures like the temporal bone, limiting its clinical efficacy.
Innovation Solution
A CT reconstruction method that combines detailed-view scanning data with full-view scanning data, utilizing a neural network for ROI projection estimation and image optimization to integrate data integrity and high resolution, improving image quality and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional CT reconstruction method is used, then reconstruction process is simple, but image quality is poor and artifacts are caused due to incomplete data
Solution Approach 1:
The patent segments the CT scanning process into two distinct modes: detailed-view scanning for high-resolution imaging of specific regions and full-view scanning for comprehensive coverage. This segmentation allows each mode to optimize for its specific purpose, with detailed-view providing high-resolution data for regions of interest and full-view providing complete anatomical context, thereby improving overall image quality while managing complexity through functional division.
Solution Approach 2:
The patent applies local quality by using detailed-view scanning with high-resolution detectors for specific regions of interest (such as temporal bone) while using full-view scanning with standard-resolution detectors for the rest of the body. This localized application of high-resolution imaging only where needed improves image quality in critical areas without unnecessarily increasing the complexity and cost of the entire scanning system.
2Measurement precision
If detailed-view scanning with high-resolution detector is used, then imaging resolution is improved, but scanning time and data processing complexity increase
Solution Approach 1:
The patent applies partial action by implementing detailed-view scanning only for specific regions of interest rather than the entire body. The system identifies regions requiring high-resolution imaging (such as temporal bone structures) and applies detailed-view scanning only to those areas, while using faster full-view scanning for the rest of the body. This partial application of high-resolution scanning maintains imaging precision where needed while significantly reducing overall scanning time and computational burden.
3Adaptability or versatility
If traditional single-mode scanning is used, then scanning process is simple, but cannot meet diagnostic requirements for fine structures
Solution Approach 1:
The patent implements universality by creating a dual-mode scanning system where a single CT scanner can perform both detailed-view scanning and full-view scanning based on diagnostic requirements. The system includes a detector module that can operate in high-resolution mode for fine structure imaging and standard-resolution mode for general imaging, allowing one device to serve multiple diagnostic purposes and adapt to different clinical scenarios without requiring separate scanning systems.
Data Source
Figure 1~2
Figure 3~4
AI summary
The present disclosure discloses a computerized tomography (CT) reconstruction method and apparatus, an electronic device and a computer-readable storage medium. The method includes: acquiring detailed-view scanning data and full-view scanning data; performing ROI projection estimation on the detailed-view scanning data to obtain detailed-view projection estimation; performing analytical reconstruction processing on the detailed-view projection estimation to obtain first image estimation of a ROI; performing reconstruction on the full-view projection data to obtain second image estimation; and determining high-resolution image of the ROI using an image optimization model according to the first image estimation and first full-view image estimation. According to the embodiments of the present disclosure, a data integrity characteristic of full-view scanning and a high resolution characteristic of detailed-view scanning can be combined to reconstruct a CT image, which not only improves the quality of a reconstructed CT image, but also greatly improves the reconstruction efficiency.