Material Decomposition Imaging with Multi-Frequency Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiation imaging systems using flat panel detectors face issues with increased noise in images obtained through energy subtraction, particularly in Interventional Radiology, which affects the visibility of contrast agents and medical devices.
Innovation Solution
An image processing apparatus that performs noise reduction processing and material decomposition by resolving material thickness images into multiple frequencies, combining noise reduction images for each frequency component, and generating images through material re-decomposition using a signal processing unit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If energy subtraction is used to reduce soft tissue or bone contrast, then the visibility of contrast agent or medical device improves, but noise in the image increases
Solution Approach 1:
The image processing is divided into multiple frequency components through Fourier transformation. Different filtering operations are applied to different frequency components: low-pass filtering to high-pass filtering across the frequency spectrum, allowing selective noise reduction while preserving important image features at each frequency level.
Solution Approach 2:
Different filtering strategies are applied to different regions of the frequency spectrum. The filtering operation varies by frequency component, with specific attention to preserving low-frequency components that contain important anatomical information while removing high-frequency noise components.
2Object-affected harmful factors
If filter processing is applied to reduce noise, then image quality improves, but important high-frequency information may be lost
Solution Approach 1:
The frequency spectrum is segmented into multiple bands, and different filtering operations are applied to each segment. This allows aggressive noise filtering in high-frequency regions while preserving or even enhancing important mid-frequency components that contain diagnostic information.
Solution Approach 2:
Instead of applying a single global filter that loses information, the patent inverts the approach by applying progressive filtering from low to high frequencies and then reconstructing the image. This reverse reconstruction process allows recovery of important information that would otherwise be lost.
Data Source
AI summary
An image processing apparatus includes: a processing unit configured to perform noise reduction processing and material decomposition processing using a plurality of images corresponding to a plurality of different radiation energies obtained by irradiating an object with radiation and performing imaging. The processing unit generates a noise reduction image by applying filter processing for each frequency component obtained by resolving a thickness image of a material obtained by the material decomposition processing into a plurality of frequencies, and generates an image by material re-decomposition processing using the thickness image after noise reduction, which is obtained by combining the noise reduction images for each frequency component and an accumulation image obtained based on combining of the plurality of images.