Low-Dose Image Denoising via Iterative Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional radiotherapy methods face challenges in generating high-quality images from low-dose scans due to noise and artifacts, which can lead to inaccurate diagnoses and increased radiation exposure for patients.
Innovation Solution
A method and system that utilize a denoising model, such as a neural network or iterative reconstruction algorithm, to process low-dose image data and enhance it to a higher equivalent dose level, improving image quality while reducing radiation exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a low-dose IGRT scan is used, then patient radiation exposure is reduced, but image quality deteriorates due to noise and artifacts
Solution Approach 1:
The patent introduces an intermediary processing system that takes low-dose scan data as input and produces enhanced image quality as output. The system uses projection data from low-dose scans, applies iterative reconstruction algorithms with noise suppression, and generates images with quality comparable to high-dose scans without directly exposing patients to high radiation doses.
Solution Approach 2:
The patent replaces the mechanical/physical approach of increasing radiation dose to improve image quality with a computational approach. Instead of using more radiation (physical mechanism), the system uses iterative mathematical reconstruction algorithms, noise modeling, and image processing techniques to enhance image quality from low-dose data.
2Measurement precision
If a high-dose IGRT scan is used, then image quality is improved, but patient radiation exposure increases
Solution Approach 1:
The patent creates a computational copy or simulation of what a high-dose scan would produce, but without actually performing a high-dose scan. By using iterative reconstruction and noise suppression algorithms on low-dose projection data, the system generates images that replicate the quality characteristics of high-dose scans while maintaining the safety advantages of low-dose exposure.
Solution Approach 2:
The patent changes the processing parameters and computational approach rather than the physical acquisition parameters. By adjusting reconstruction algorithm parameters, noise model parameters, and iteration counts, the system transforms low-dose projection data into high-quality images, effectively decoupling image quality from radiation dose parameters.
3Device complexity
If conventional reconstruction methods are used on low-dose data, then processing is simple, but image quality suffers from noise and artifacts
Solution Approach 1:
The patent performs preliminary actions in the reconstruction process by pre-modeling noise characteristics, pre-defining iteration parameters, and pre-establishing reconstruction algorithms tailored for low-dose data. This preliminary preparation enables the system to systematically address noise and artifacts before they degrade image quality, allowing complex processing to be managed through structured preliminary steps.
Solution Approach 2:
The patent implements continuous iterative reconstruction where the useful action of image refinement continues through multiple iterations. Each iteration progressively improves image quality by suppressing noise and correcting artifacts, maintaining continuous refinement until convergence criteria are met, rather than using a single-step reconstruction that would leave images degraded.
Data Source
AI summary
A method may include obtaining first image data relating to a region of interest (ROI) of a first subject. The first image data corresponding to a first equivalent dose level may be acquired by a first device. The method may also include obtaining a model for denoising relating to the first image data and determining second image data corresponding to an equivalent dose level higher than the first equivalent dose level based on the first image data and the model for denoising. In some embodiments, the method may further include determining information relating to the ROI of the first subject based on the second image data and ecording the information relating to the ROI of the first subject.