Radiation Image Noise Reduction via Adaptive Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radiation imaging systems face challenges in reducing noise, particularly photon noise, which degrades image quality and introduces artifacts such as streaking and blurring, especially when lower radiation doses are used, as the noise level is inversely related to the dose applied.
Innovation Solution
A method and system for processing images from radiation examinations that estimate a target noise contribution, filter the images to generate a filtered image, and iteratively compare the noise contributions to ensure the target noise level is met, combining the filtered image with the original image when the target is satisfied, using techniques like outlier filtering and diffusion filtering to reduce noise while preserving image features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If radiation dose is decreased to reduce patient exposure, then safety improves, but photon noise increases and image quality deteriorates
Solution Approach 1:
The patent converts the harmful effect of photon noise into a beneficial filtering process. By estimating noise contributions from different sources (Poisson noise from photon statistics, Gaussian noise from detector electronics) and applying targeted filtering, the system removes noise while preserving image quality, thereby allowing lower radiation doses without sacrificing diagnostic value
Solution Approach 2:
The patent changes the statistical parameters used for noise estimation and filtering. It models noise as a combination of Poisson and Gaussian distributions with different variances, and uses these parameter estimates to adaptively filter the image. This parameter-based approach enables effective noise reduction at low doses by accurately characterizing and removing noise components
2Measurement precision
If filtering is applied to reduce noise, then image quality improves, but excessive smoothing may occur and obscure important image features
Solution Approach 1:
The patent applies local quality by using adaptive filtering that adjusts filtering strength based on local image characteristics. The filter estimates noise parameters locally and applies appropriate filtering only where needed, preserving edges and important features while reducing noise in homogeneous regions. This localized approach prevents excessive smoothing of diagnostically important structures
Solution Approach 2:
The patent implements feedback by iteratively estimating noise contributions, applying filters, re-estimating noise, and comparing the results to determine whether the target noise level has been satisfied. This feedback loop allows the system to stop filtering when the desired noise level is achieved, preventing over-filtering and preservation of image features
Data Source
AI summary
Among other things, one or more techniques and/or systems are described for processing images yielded from an examination via radiation to reduce visible noise in the images. After an image is reconstructed, a noise contribution to the image (e.g., an amount of noise in the image) is estimated to determine a target noise contribution for the image. The target noise contribution for the image may vary based upon, among other things, dose of radiation, aspects or properties of an object being imaged, etc. The image is subsequently filtered using one or more filtering techniques to generate a filtered image, and a noise contribution to the filtered image is determined. When the noise contribution to the filtered image satisfies the target noise contribution (e.g., a sufficient amount of noise has been filtered out of the image), the filtered image is combined with the reconstructed image to generate a blended image.