AI Calcification Detection in Angiography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for treating coronary artery disease (CAD) with calcified plaque are associated with higher complications and lower success rates compared to non-calcified plaque, particularly due to stent under-expansion during percutaneous coronary intervention (PCI), necessitating an effective way to assess and quantify coronary artery calcification.
Innovation Solution
Training an artificial intelligence model to detect calcified portions in medical images by co-registering and projecting calcified features from one modality to another, such as from computed tomography (CT) to x-ray angiography, or generating synthesized images to enhance detection accuracy across different imaging modalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PCI is performed on calcified plaque without proper assessment, then the procedure can be completed, but stent under-expansion occurs leading to higher complications and lower success rates
Solution Approach 1:
The system performs preliminary detection and quantification of coronary calcification using AI analysis of angiographic images before PCI procedure. This allows clinicians to assess calcification burden in advance and plan appropriate debulking strategies (such as rotational atherectomy) before stent deployment, preventing stent under-expansion and improving procedural success rates
Solution Approach 2:
The AI model serves as an intermediary tool that bridges the gap between standard angiographic imaging and accurate calcification assessment. By training the AI on multi-modal images (CT, IVUS, OCT) and applying it to routine angiography, the system enables non-invasive calcification quantification without requiring additional invasive imaging procedures during the PCI procedure
2Reliability
If rotational atherectomy is performed to ablate calcified plaque, then procedure success improves, but total procedure time increases
Solution Approach 1:
The AI system performs calcification assessment during the routine angiographic imaging phase, which is already part of the PCI workflow. By providing real-time or near-real-time calcification analysis without requiring separate imaging procedures or significant additional processing time, the system enables timely clinical decision-making about debulking necessity, allowing procedure planning to occur seamlessly within the existing workflow timeline
3Measurement precision
If multiple imaging modalities are used to assess calcification, then detection accuracy improves, but device complexity and cost increase
Solution Approach 1:
The system creates a virtual model of calcification appearance by training an AI model on multi-modal images (CT, IVUS, OCT) that show calcified plaque. This trained model then serves as a virtual expert that can analyze and detect calcification in routine angiographic images, replicating the diagnostic accuracy of multiple imaging modalities using only a single modality (angiography) during the actual procedure
Solution Approach 2:
The AI model learns to detect calcification by analyzing changes in image parameters (intensity, texture, edge characteristics) that differentiate calcified from non-calcified plaque. By training on multi-modal data, the model becomes sensitive to subtle parameter variations in angiographic images that indicate calcification, achieving high detection accuracy through sophisticated parameter analysis rather than additional imaging hardware
Data Source
AI summary
Systems and methods are provided for training an artificial intelligence model for detecting calcified portions of a vessel in an input medical image. One or more first medical images of a vessel in a first modality and one or more second medical image of the vessel in a second modality are received. Calcified portions of the vessel are detected in the one or more first medical images, The artificial intelligence model is trained for detecting calcified portions of the vessel in the input medical image in the second modality based on the one or more second medical images and the detected calcified portions of the vessel detected in the one or more first medical images.