Dynamic Lung CT Image Concatenation via Motion-Matched Projection Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional X-ray CT imaging techniques face challenges in capturing the entire pulmonary field dynamically due to the size limitations of area detectors, requiring multiple scans and precise matching of breathing depths, which can be difficult to achieve.
Innovation Solution
A medical image processing apparatus that obtains projection data for different lung regions along with dynamic data representing subject motion, extracts projection datasets based on matching dynamic data thresholds and motion magnitudes, and reconstructs volume data for concatenation, enabling the generation of dynamic images across larger areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If dynamic imaging is performed twice or more to capture the entire pulmonary field, then the imaging area is improved, but the difficulty of matching breathing depth and speed increases
Solution Approach 1:
The pulmonary field is divided into multiple regions (first region and second region) that are imaged separately and then concatenated. This segmentation allows the use of a smaller area detector to capture the entire pulmonary field by performing multiple scans, while the processing apparatus automatically handles the concatenation without requiring manual breathing synchronization.
Solution Approach 2:
The processing apparatus automatically determines whether to concatenate volume data by comparing dynamic data (breathing information) from multiple scans without requiring manual intervention. The system self-evaluates the suitability of concatenation based on matching breathing depth and speed, eliminating the need for operators to manually synchronize breathing across multiple scans.
2Device complexity
If a smaller area detector is used, then the device complexity is reduced, but the imaging area decreases
Solution Approach 1:
Instead of using a single large-area detector, the system uses a smaller area detector to capture multiple regions sequentially. The pulmonary field is segmented into first and second regions that are imaged in separate scans and then concatenated computationally, achieving full-field coverage with simpler, smaller detector hardware.
Solution Approach 2:
The system transitions from spatial coverage (using a large detector) to temporal coverage (performing multiple scans at different time points). By imaging the same pulmonary field at different times and concatenating the results, the system achieves full-field coverage without requiring a large-area detector, effectively moving the solution from the spatial domain to the temporal domain.
3Area of stationary object
If multiple scans are performed to capture the entire pulmonary field, then the imaging area is improved, but the time required increases
Solution Approach 1:
The system performs preliminary evaluation of dynamic data (breathing information) from multiple scans to determine whether they are suitable for concatenation before actually performing the concatenation operation. This preliminary assessment prevents wasted time on incompatible scans and streamlines the workflow by quickly identifying suitable candidate scans for concatenation.
Solution Approach 2:
The processing apparatus automatically evaluates and determines the suitability of multiple scans for concatenation based on dynamic data comparison, eliminating the need for manual review and selection. This automated self-evaluation process reduces the time required to identify suitable scans for concatenation and streamlines the overall imaging workflow.
Data Source
AI summary
A medical image processing apparatus includes processing circuitry that obtains 1st projection data of a 1st region with 1st dynamic data and 2nd projection data of a 2nd region with 2nd dynamic data; when the 1st and 2nd dynamic data exceed a threshold and match in magnitude of motion, extracts, from the 1st and 2nd projection dataset for an angular range at a time interval or for the magnitude, a 1st projection dataset and 2nd projection dataset, respectively; when the 1st and 2nd dynamic data exceed the threshold and differ in the magnitude, extracts the 1st projection dataset from the 1st projection data for the magnitude, and extracts the 2nd projection dataset for the same magnitude from the 2nd projection data; reconstructs 1st and 2nd volume data based on the 1st and 2nd projection datasets, respectively; and generates volume data concatenating the 1st and 2nd volume data.