Automatic Agatston Score Computation from Coronary CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual quantification of the Agatston score from coronary computed tomography (CT) images is tedious and error-prone, and requires both CT and CT angiography (CTA), increasing cost and radiation exposure.
Innovation Solution
An automatic method using joint atlas label fusion for comprehensive spatial information and feature extraction from CT images alone, employing a random forest classifier to detect coronary calcification and compute the Agatston score, reducing reliance on CTA and minimizing radiation exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual quantification of Agatston score is performed using both CT and CTA, then measurement precision is improved, but loss of time and loss of energy increase
Solution Approach 1:
The system performs automatic self-quantification of Agatston scores using machine learning algorithms that autonomously analyze CT images, extract calcium deposits, calculate scores, and generate reports without requiring manual radiologist intervention, thereby eliminating time loss while maintaining precision
Solution Approach 2:
The patent replaces the manual mechanical process of radiologist measurement with an automated computer-based system using deep learning convolutional neural networks that automatically detect calcium deposits and compute Agatston scores, significantly reducing time consumption while preserving measurement accuracy
2Measurement precision
If both CT and CTA are used for Agatston score quantification, then measurement precision is improved, but use of energy and cost increase
Solution Approach 1:
The patent extracts and utilizes only the essential information needed for Agatston score calculation from standard CT images, eliminating the need for additional CTA scans. The machine learning model is trained to identify calcium deposits and compute scores from routine CT data alone, thereby reducing radiation exposure while maintaining quantification precision
Solution Approach 2:
The system enables standard CT scans to serve multiple purposes: both routine anatomical imaging and automatic Agatston score quantification. The multi-functional AI platform processes CT images to simultaneously provide diagnostic information and calcium scoring, eliminating the need for separate CTA procedures and reducing overall radiation exposure
3Measurement precision
If manual quantification is performed, then measurement precision is maintained, but productivity decreases
Solution Approach 1:
The automated system performs complete self-quantification including image analysis, calcium detection, score calculation, and report generation without manual intervention, enabling high-throughput processing of multiple patient scans simultaneously while maintaining precision through validated machine learning algorithms
Solution Approach 2:
The patent replaces slow manual radiologist measurement with high-speed automated computer processing using convolutional neural networks that can analyze and quantify calcium scores in seconds per patient, dramatically increasing productivity and throughput while maintaining or improving measurement precision through consistent algorithmic application
Data Source
AI summary
Automatic characterization of the Agatston score from coronary computed tomography (CT) is provided. In various embodiments, a plurality of coronary computed tomography images are segmented into a plurality of segments corresponding to features of coronary anatomy. A plurality of calcium candidates are extracted from the plurality of coronary computed tomography images by thresholding. Coronary calcification is located in the coronary computed tomography images by applying a trained classifier to the plurality of calcium candidates. An Agatson score is computed from the located calcification.