Tomosynthesis Gain Map Correction Using Angle-Specific Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gain maps used in mammography are inadequate for tomosynthesis imaging due to the greater number of x-ray source projection angle changes and lower x-ray exposure in tomosynthesis, leading to suboptimal image correction.
Innovation Solution
A tomosynthesis system acquires x-ray projection images of a calibration phantom at multiple angles to generate initial gain maps, which are then combined to create enhanced gain maps for correcting pixel value differences, allowing for accurate gain correction of breast images taken at varying projection angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional gain maps are used for tomosynthesis imaging, then the system can operate with existing calibration data, but the image correction is suboptimal due to inadequate calibration for multiple projection angles
Solution Approach 1:
The gain map generation process is segmented into multiple angle-specific gain maps, where each gain map corresponds to a specific projection angle. This allows for precise correction at each angle while maintaining overall system manageability through modular processing of angle-specific data.
Solution Approach 2:
The calibration process is extended from a single-plane (2D) conventional gain map to a multi-angle three-dimensional gain map structure. This adds the angular dimension to the calibration data, enabling accurate correction across multiple projection angles while organizing the complexity in a structured dimensional framework.
2Measurement precision
If multiple angle-specific gain maps are generated for tomosynthesis, then the Contrast Noise Ratio (CNR) is improved, but the processing complexity increases
Solution Approach 1:
Angle-specific gain maps are pre-calculated and stored for each projection angle during system calibration. This preliminary action eliminates the need for real-time complex calculations during tomosynthesis imaging, as the correction factors are already prepared and can be applied directly to each projection angle.
Solution Approach 2:
The system changes the calibration parameter from a single uniform gain map to multiple angle-specific gain maps with different correction parameters. This allows optimization of CNR for each angle while the structured parameter organization enables efficient retrieval and application during imaging.
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
The enhanced gain maps improve the Contrast Noise Ratio (CNR) of tomosynthesis images and reduce the x-ray dose penalty, resulting in higher quality, uniformly corrected breast images.
Implementation Method 1
acquires a multiplicity of x-ray projection images T(p,n) of a calibration phantom, where p designates a respective one of a multiplicity of P first projection angles of an x-ray beam
Data Source
AI summary
A multi-mode tomosynthesis/mammography system and method in which a mammography gain map is used to gain correct mammographic images of a patient's breast but enhanced gain maps for respective projection angled are used to correct tomosynthesis images acquired with the same system.