Energy-Resolved Image Filtering for Fluoroscopic Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In fluoroscopic radiography using Flat Panel Detectors, reducing radiation dose increases quantum noise in images, and existing technologies do not effectively address this noise reduction.
Innovation Solution
An image processing apparatus that filters signal components obtained from energy images at different energy levels, generating a moving image with distinct filter coefficients for each component to improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If radiation dose is reduced to minimize patient exposure, then radiation safety is improved, but quantum noise in images increases
Solution Approach 1:
The image signal is segmented into multiple energy components (first and second signal components corresponding to different energy levels). By separating and independently processing these energy components, the system can apply different filtering strategies to each component, thereby reducing noise while preserving diagnostic information and maintaining image quality even at low radiation doses.
Solution Approach 2:
The patent applies different filter coefficients to different energy components based on their respective noise characteristics. By dynamically adjusting filtering parameters (filter coefficients) according to the energy level and noise properties of each component, the system optimizes noise reduction while maintaining image fidelity, effectively resolving the contradiction between low dose and image quality.
2Loss of information
If energy subtraction processing is performed to divide images into bone and soft tissue components, then diagnostic information is improved, but noise in the divided images increases
Solution Approach 1:
The patent segments the image into multiple energy components that correspond to different tissue types (bone and soft tissue). By maintaining separate signal components for different energy levels throughout the processing pipeline, the system preserves diagnostic information while enabling independent noise management for each tissue type, thus reducing noise in the final divided images.
Solution Approach 2:
Different filter coefficients are applied to different energy components based on their local characteristics. The first signal component (lower energy) and second signal component (higher energy) receive tailored filtering that considers their specific noise properties, thereby reducing noise in the divided images while preserving the diagnostic information unique to each tissue type.
3Measurement precision
If filtering is applied to reduce noise in signal components, then image quality is improved, but image lag increases
Solution Approach 1:
The patent dynamically adjusts filter coefficients based on the temporal characteristics of the imaging sequence. By adapting the filtering strength according to the specific requirements of each energy component and the temporal dynamics of the fluoroscopic sequence, the system achieves noise reduction while minimizing image lag, as the filtering parameters are optimized to balance quality and temporal responsiveness.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An image processing apparatus comprises a filtering unit for performing recursive filtering on a first signal component and a second signal component that are obtained by emitting radiation at a plurality of levels of energy toward an object, and a generation unit for generating a moving image based on the first signal component and the second signal component on which the recursive filtering is performed. A filter coefficient of the recursive filtering performed on the first signal component and a filter coefficient of the recursive filtering performed on the second signal component differ from each other.