Motion-Corrected CT Reconstruction via Partial Angle Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical imaging technologies, such as CT scans, often produce blurred images due to patient movement during scanning, leading to motion artifacts that complicate accurate diagnosis.
Innovation Solution
A medical image processing apparatus and method that generates partial angle reconstruction (PAR) images, performs correspondence mapping based on characteristics of these images, and produces a motion-corrected reconstruction image using correspondence-mapping data and scan data to align and correct for motion artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional CT scanning is used to capture images of a subject, then scan data can be obtained, but motion artifacts and blurred images are produced due to patient movement during scanning
Solution Approach 1:
The patent divides the complete CT scan into multiple partial angle reconstruction (PAR) images, each representing a specific angular range. This segmentation allows the system to process and align different portions of the scan independently, enabling motion correction by identifying and correcting misalignments between PAR images without requiring the entire scan to be perfect.
Solution Approach 2:
The patent introduces a correspondence mapping dimension that maps PAR images from different angular ranges into a common reference frame. By creating correspondence-mapping data that describes relationships between PAR images, the system adds a dimensional layer for alignment and registration, transforming the problem from simple image reconstruction to multi-dimensional spatial alignment.
2Measurement precision
If scanning time is prolonged to improve image quality, then motion artifacts can be reduced, but diagnostic time and patient exposure increase
Solution Approach 1:
The patent performs preliminary processing by generating multiple PAR images from the scan data before final reconstruction. These PAR images serve as intermediate representations that can be pre-aligned and registered. By preparing and aligning PAR images in advance, the system can correct motion artifacts without requiring extended scanning time, as the alignment is performed on processed data rather than raw scan data.
Solution Approach 2:
The patent creates multiple copies of the scan data in the form of PAR images, each representing a specific angular perspective. These copies are then used to generate correspondence-mapping data that describes spatial relationships. By working with these replicated angular views, the system can perform motion correction algorithms that would otherwise require time-consuming full-scenarion re-scanning.
3Manufacturing precision
If multiple PAR images are generated and processed to create motion-corrected reconstruction, then image alignment accuracy improves, but processing complexity increases
Solution Approach 1:
The patent segments the complex processing task into distinct stages: generating PAR images, obtaining characteristics, performing correspondence mapping, and creating the final motion-corrected reconstruction. Each stage processes a specific subset of data and produces intermediate results, making the overall complex process more manageable and systematic.
Solution Approach 2:
The patent introduces correspondence-mapping data as an intermediary that mediates between the PAR images and the final reconstruction. This intermediary layer captures the spatial relationships and alignment information between PAR images, allowing the final reconstruction to be generated without directly processing all the complex relationships between every pair of PAR images simultaneously.
Data Source
AI summary
Devices, systems, and methods receive scan data that were generated by scanning a region of a subject with a computed tomography apparatus; generate multiple partial angle reconstruction (PAR) images based on the scan data; obtain corresponding characteristics of the multiple PAR images; perform correspondence mapping on the multiple PAR images based on the obtained corresponding characteristics and on the multiple PAR images, wherein the correspondence mapping generates correspondence-mapping data; and generate a motion-corrected reconstruction image based on the correspondence-mapping data and on one or both of the scan data and the PAR images.