Tomography Image Reconstruction Using Noise-Reduced Reference Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current tomography apparatuses face challenges in reconstructing noise-free cross-sectional images, particularly due to motion artifacts and noise interference, which affect the accuracy of medical diagnoses.
Innovation Solution
The proposed solution involves a tomography apparatus that obtains first and second image data points using X-ray irradiation, performs noise reduction on these data sets using techniques such as low pass filtering, edge region detection, and voxel/slice adjustments, and reconstructs a target image based on noise-reduced reference images, while considering X-ray dose and irradiation conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional tomography reconstruction is performed without noise reduction, then the reconstruction process is simple and fast, but the resulting cross-sectional image contains noise and motion artifacts that reduce diagnostic accuracy
Solution Approach 1:
The patent applies noise reduction processing to obtain reference images before performing the final tomography reconstruction. By preprocessing the image data to reduce noise and correct motion artifacts in advance, the final reconstruction uses cleaner reference images, improving overall image quality without significantly increasing the complexity of the reconstruction algorithm itself.
2Measurement precision
If noise reduction processing is applied to image data, then the cross-sectional image quality improves, but the processing time and computational complexity increase
Solution Approach 1:
The patent applies noise reduction processing selectively to specific regions or components of the image data that most benefit from it, rather than uniformly processing all data. This partial application of noise reduction achieves sufficient image quality improvement while reducing the overall computational burden and processing time compared to exhaustive noise reduction of the entire dataset.
3Measurement precision
If multiple reference images are used for reconstruction, then the accuracy of motion correction improves, but the complexity of image processing and reconstruction increases
Solution Approach 1:
The patent combines multiple reference images obtained from different time points or processing conditions to create a composite reference image for reconstruction. By merging these multiple images, the system achieves better motion correction accuracy as the combined reference contains more comprehensive information about object motion, while the merging process itself provides a systematic approach that manages the complexity of handling multiple images.
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
This approach enables the reconstruction of high-quality, noise-free cross-sectional images that minimize motion artifacts, thereby enhancing diagnostic accuracy and image clarity.
Implementation Method 1
raw data is obtained by performing tomography on an object via a CT apparatus... obtained by irradiating an X-ray to an object
Data Source
AI summary
A tomography apparatus includes a data obtainer configured to obtain first image data at a first point and second image data at a second point using tomography, the tomography being performed by irradiating an X-ray to an object; an image processor configured to perform noise reduction based on at least one from among the first image data and the second image data, and to obtain a first reference image corresponding to the first image data and a second reference image corresponding to the second image data using a result of the performed noise reduction; and an image reconstructor configured to reconstruct a target image representing the object based on the first reference image and the second reference image.