CTA-Registered CT Thrombosis Detection via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical imaging technologies, such as CT and MRI, face challenges in accurately identifying and visualizing cerebral thrombosis due to difficulty in distinguishing thrombosis from adjacent tissues, and limitations in CTA imaging, including increased patient exposure to radiation and side effects from contrast agents.
Innovation Solution
A computing system and method that registers CT and CTA images to identify candidate blood vessel regions using a specific threshold value for brightness, generates an average standard model from CTA images to create a mask layer, and visualizes potential thrombosis locations by distinguishing them from other image portions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional CT imaging is used to search for cerebral thrombosis, then the entire cerebral region must be examined, but this increases the time and complexity of diagnosis due to difficulty in distinguishing thrombosis from adjacent tissues
Solution Approach 1:
The patent segments the cerebral region into blood vessel regions and non-blood vessel regions by registering CT images with CTA images. This segmentation allows the system to focus thrombosis detection only within blood vessel regions, reducing the search space and diagnosis time while maintaining detection accuracy.
Solution Approach 2:
The patent performs preliminary registration of CT images with CTA images to identify blood vessel regions before conducting thrombosis detection. This preliminary action of defining the search region upfront eliminates the need to examine the entire cerebral region, thereby reducing diagnosis time without compromising detection accuracy.
2Measurement precision
If CTA imaging is used to enhance blood vessel visualization, then thrombosis location can be more accurately identified, but patient exposure to radiation and contrast agents increases
Solution Approach 1:
The patent merges CT imaging (which has lower radiation dose and no contrast agent) with CTA imaging (which provides excellent blood vessel visualization) by registering the two image sets. This combination allows the system to use the low-dose CT for general imaging while leveraging the CTA for accurate blood vessel localization, thereby reducing overall radiation exposure and contrast agent usage while maintaining thrombosis detection accuracy.
3Reliability
If the entire cerebral region is searched for thrombosis in CT images, then no region is missed, but the complexity of image analysis increases due to the large search space
Solution Approach 1:
The patent segments the large cerebral region into smaller blood vessel regions using CTA image registration. This segmentation reduces the search space from the entire cerebral region to only the relevant blood vessel regions, thereby reducing image analysis complexity while maintaining detection completeness through systematic coverage of all segmented regions.
Solution Approach 2:
The patent applies different processing strategies to different regions: CTA-based blood vessel segmentation is applied to identify search regions, while threshold-based thrombosis detection is applied specifically within those regions. This local quality approach optimizes the analysis process for each region's characteristics, reducing overall complexity while ensuring thorough detection.
Data Source
AI summary
Disclosed are a computing system and method for identifying and visualizing a thrombosis. The method includes: registering an acquired computer tomography (CT) image of an examinee and a computer tomography angiography (CTA) image-based average standard model to each other; identifying a candidate blood vessel region from the CT image based on the results of the registration; acquiring a volume, selected by applying a threshold value for the brightness of the CT image to the candidate blood vessel region, as a first thrombosis candidate region; acquiring at least part of the first thrombosis candidate region as a second thrombosis candidate region based on the size information of the first thrombosis candidate region; and visualizing the second thrombosis candidate region by assigning a visual effect, adapted to distinguish the second thrombosis candidate region from the remaining portion of the CT image, to the second thrombosis candidate region.