CT Projection-Domain Correction for Scatter and Detector Crosstalk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CT image quality is degraded by detector cross-talk, object scatter, and background scatter, which introduce artifacts, reduce spatial resolution, and degrade signal-to-noise ratio, often requiring costly hardware and frequent supplemental scans.
Innovation Solution
A software-based correction method using pre-calculated models for object scatter, background scatter, and detector crosstalk, applied in the projection domain to compensate for these effects without hardware-based measurements, calibrated on a per material/spectrum and per scanner basis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hardware-based scatter profile measurement is used, then scatter correction accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces hardware-based scatter profile measurement systems with a software-based correction method. Instead of using physical hardware components to measure and correct scatter, the system uses pre-calculated lookup tables and mathematical models processed through software algorithms, thereby reducing hardware complexity while maintaining correction accuracy.
Solution Approach 2:
The patent creates a virtual copy of the scatter correction process through pre-calculated lookup tables that store correction data. These lookup tables are generated in advance and stored in memory, allowing the system to retrieve and apply correction factors without requiring complex real-time hardware measurements, thus simplifying the physical system while preserving measurement precision.
2Measurement precision
If supplemental scans are performed for correction, then image quality is improved, but loss of time increases
Solution Approach 1:
The patent performs scatter correction calculations in advance by pre-calculating and storing lookup tables during system setup or calibration. This preliminary action eliminates the need for time-consuming supplemental scans during actual imaging, as the correction data is already ready to be applied instantly to the projection data, thereby improving image quality without adding scan time.
Solution Approach 2:
The patent segments the correction process into separate pre-calculated components stored in lookup tables. By dividing the correction task into pre-computed segments that can be independently stored and retrieved, the system avoids performing complete correction calculations in real-time during scanning, thus maintaining image quality while reducing the time required for correction operations.
3Reliability
If hardware-based correction systems are used, then scatter correction capability is improved, but device complexity and cost increase
Solution Approach 1:
The patent substitutes hardware-based correction mechanisms with software-based processing. Instead of using physical hardware systems to perform scatter correction, the invention uses software algorithms that process projection data through pre-calculated lookup tables, achieving reliable scatter correction with significantly reduced system complexity and lower cost.
Solution Approach 2:
The patent implements a self-service correction approach where the system uses its own pre-calculated lookup tables to perform correction without requiring external hardware intervention. The system self-corrects scatter effects by retrieving appropriate correction factors from stored tables and applying them to the projection data, thereby maintaining correction capability while minimizing system complexity.
Data Source
Figure 1~2
Figure 3~5
Figure 6
AI summary
A control circuit (101) accesses (201) a memory (102) having stored therein total detected computed tomography image signal information in a projection domain for an object, a first model representing object scatter, a second model representing background scatter, and a third model representing detector crosstalk. Prior to a reconstruction step, the control circuit processes (202) the total detected computed tomography image signal information in the projection domain for the object as a function of each of the first model, the second model, and the third model to thereby compensate for object scatter, background scatter, and detector crosstalk by providing scatter and crosstalk-corrected computed tomography image signal information in the projection domain for the object. The control circuit can then carry out a reconstruction step (203) using the scatter and crosstalk-corrected computed tomography image signal information in the projection domain for the object.