CT Angiography Calcium Removal for Clearer Arterial Lumens
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for removing calcium artifacts in cardiac computed tomography angiography images are inadequate, leading to inaccurate analysis of arterial lumens and potential overestimation of cardiac diseases due to obscured vessel structures.
Innovation Solution
A method involving neural networks, such as convolutional neural networks with contracting and expanding paths, and a sliding window technique, is used to identify and remove calcium deposits by generating calcium-free image patches, enhancing the visibility of arterial lumens in CT images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional subtraction or segmentation methods are used to remove calcium artifacts, then some calcium removal is achieved, but the precision of calcium artifact removal is insufficient leading to inaccurate arterial lumen analysis
Solution Approach 1:
The patent segments the image processing task into multiple specialized neural network components: a calcium detection network that identifies calcium deposits, an inpainting mask generation module that creates removal masks, and an image restoration network that reconstructs the arterial lumen. This segmentation allows each component to specialize in one aspect of the problem, improving overall precision compared to traditional single-step subtraction or segmentation methods.
Solution Approach 2:
The patent introduces an intermediary inpainting mask as a bridge between calcium detection and image restoration. The mask serves as an intermediate representation that precisely delineates calcium regions before restoration, enabling more accurate removal than direct subtraction methods. This intermediary step allows for controlled, precise calcium artifact removal while preserving arterial lumen integrity.
2Reliability
If calcium artifacts are not removed, then image acquisition is simple and fast, but calcium deposits obscure arterial lumens leading to overestimation of cardiac diseases
Solution Approach 1:
The patent implements self-service through automated neural network-based calcium detection and removal. The system automatically identifies calcium deposits, generates removal masks, and restores arterial lumen images without requiring manual intervention or complex multi-step processing protocols. This automation maintains high diagnosis accuracy while managing processing complexity through integrated AI algorithms.
Solution Approach 2:
The patent replaces traditional mechanical image processing methods (subtraction algorithms, manual segmentation) with neural network-based computational approaches. The deep learning models automatically learn to identify and remove calcium artifacts, substituting complex mechanical processing workflows with more efficient AI-driven methods that maintain diagnostic reliability.
3Productivity
If manual segmentation methods are used to remove calcium, then some artifact removal is achieved, but the process is time-consuming and reduces productivity
Solution Approach 1:
The patent employs periodic action through the automated, iterative operation of neural networks that rapidly process images through predefined stages: calcium detection, mask generation, and image restoration. This automated periodic processing replaces time-consuming manual segmentation while maintaining high productivity through efficient algorithm execution.
Solution Approach 2:
The system performs self-service by automatically executing the complete calcium removal workflow without manual intervention. The neural networks autonomously detect calcium, generate removal masks, and restore images, eliminating the time loss associated with manual segmentation while maintaining high processing productivity through automated AI algorithms.
Data Source
Figure 1A~2B
Figure 3
Figure 4A~4C
AI summary
A method of analyzing computed tomography (CT) images comprises receiving an initial CT image of an object, identifying calcium-free regions and a calcified region in the initial CT image of the object, generating a calcium-free image patch, and applying the calcium-free image patch to the initial CT image patch to produce a final CT image. The initial CT image shows a calcium deposit and a target structure in the object. The calcified region in the initial CT image shows the calcium deposit in the object obscuring a portion of the target structure. The calcium-free regions show the remaining portions of the target structure. The calcium-free image patch corresponds to the calcified region in the initial CT image. The final CT image shows the calcium-free image patch and the calcium-free region from the initial CT image. The calcium-free image patch is generated and applied using a convolutional neural network.