CTA Vessel Rendering with Machine Learning and Low-Concentration Contrast
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computed tomography angiography (CTA) methods face challenges with beam hardening, double covering, and beam hardening, leading to inaccurate and unreliable angiographic images due to the use of iodinated contrast agents, which can cause side effects and limit contrast enhancement.
Innovation Solution
A method using machine learning based on a calibration phantom and computer simulation to generate 3D rendering images with low-concentration contrast agents, overcoming beam hardening and double covering by training models with varied blood vessel thicknesses and medical procedure data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iodinated contrast agents are used in CTA, then contrast enhancement of blood vessels is improved, but side effects and toxicity increase
Solution Approach 1:
The patent changes the concentration parameter of the contrast agent by using low-concentration contrast media (e.g., 5-50% iodine concentration) instead of traditional high-concentration agents, thereby reducing toxicity while maintaining adequate contrast enhancement through optimized imaging protocols
Solution Approach 2:
The patent employs composite imaging techniques combining multiple imaging modalities (CTA, MRA, DSA) and processing methods (deep learning algorithms, iterative reconstruction) to achieve superior contrast enhancement without relying solely on high-concentration iodinated contrast agents
2Measurement precision
If contrast agents are used with X-ray radiation, then blood vessel contrast is improved, but DNA damage increases
Solution Approach 1:
The patent applies partial action by using low-concentration contrast agents (excessive in terms of safety margin) combined with advanced imaging algorithms that can achieve diagnostic quality images with reduced contrast agent dosage, thereby minimizing DNA damage while maintaining measurement precision
Solution Approach 2:
The patent creates virtual copies of blood vessel structures through deep learning algorithms and image processing techniques, allowing diagnostic information to be extracted from synthetic images rather than requiring high-dose contrast agents for every diagnostic question
3Measurement precision
If high-density surgical information is extracted from CTA, then diagnostic accuracy is improved, but beam hardening causes information loss
Solution Approach 1:
The patent replaces traditional mechanical image processing with deep learning-based computational methods that can distinguish between beam hardening artifacts and genuine high-density surgical structures, enabling accurate extraction of surgical information without information loss
Solution Approach 2:
The patent introduces deep learning algorithms as intermediaries between the raw CTA images and the final diagnostic output, where the AI model learns to compensate for beam hardening effects and extract accurate information about high-density materials and surgical structures
4Quantity of substance
If dual energy CTA is used to reduce contrast agent, then contrast agent amount is reduced, but development cost and complexity increase
Solution Approach 1:
The patent makes the imaging system multi-functional by integrating deep learning algorithms that can process standard single-energy CTA images to reduce contrast agent requirements, eliminating the need for expensive dual-energy equipment while achieving contrast agent reduction
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables accurate extraction of blood vessel and medical procedure data from single-imaged CTA projections, reducing side effects and improving diagnostic accuracy by generating high-precision 3D rendering images.
Implementation Method 1
A method using machine learning based on a calibration phantom and computer simulation to generate 3D rendering images with low-concentration contrast agents
Implementation Method 2
the CTA mainly uses iodinated contrast agents to increase the contrast of blood vessels
Data Source
AI summary
In the present specification, as an information providing method for angiography which is implemented by a processor, an information providing method for angiography including a step of receiving a projection image of a computed tomography angiography (CTA) of an individual, a step of generating a first extraction image for a blood vessel by inputting the received CTA projection image to a first model which is trained to generate a first extraction image for the blood vessel with a CTA projection image as an input, and a step of generating a rendering image from the first extraction image.