Elastogram Pre-segmentation for Plaque Rupture Risk Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the rupture of atheromatous plaques are imprecise due to insufficient quantification of the thickness of the fibrous cap, morphology, and mechanical properties, particularly elasticity, which are challenging to identify in vivo due to the heterogeneity and changing nature of the plaque components.
Innovation Solution
An image processing method that receives an elastogram showing internal deformations, performs pre-segmentation considering the tissue as an incompressible elastic solid, and computes an elasticity image using a space-dependent vector to estimate the risk of plaque rupture by iteratively selecting and optimizing zones, allowing for precise measurement of the fibrous cap thickness and plaque components' dimensions and mechanical properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative optimization algorithms are used to reconstruct plaque elasticity, then the accuracy of elasticity estimation is improved, but the computational complexity and processing time increase significantly
Solution Approach 1:
The patent applies preliminary segmentation to divide the plaque into distinct regions (fibrous cap, necrotic core, calcium deposits) before elasticity reconstruction. This pre-processing step simplifies the subsequent optimization problem by reducing the search space and providing initial estimates for different tissue types, thereby maintaining accuracy while reducing computational complexity.
Solution Approach 2:
The plaque is segmented into multiple homogeneous regions with distinct mechanical properties. By treating each segment separately with appropriate elasticity constraints, the overall reconstruction problem becomes more tractable while preserving local accuracy, resolving the contradiction between precision and complexity.
2Ease of manufacture
If the plaque is treated as a homogeneous material, then the analysis is simpler, but the prediction accuracy deteriorates due to ignoring the heterogeneity of plaque components
Solution Approach 1:
The plaque is divided into multiple segments (fibrous cap, necrotic core, calcium deposits, fibro-fatty tissue) each assigned distinct elasticity values. This segmentation maintains analytical tractability while capturing the heterogeneity necessary for accurate rupture risk prediction, resolving the contradiction between simplicity and precision.
Solution Approach 2:
Different regions of the plaque are assigned different material properties and elasticity values according to their specific tissue composition. This local differentiation allows the analysis to remain relatively simple while significantly improving prediction accuracy by accounting for the varying mechanical properties of different plaque components.
3Measurement precision
If advanced optimization algorithms are implemented to extract elastic moduli, then the elasticity image quality is improved, but the method fails to identify calcium inclusions and accurately quantify fibrous cap thickness
Solution Approach 1:
The patent performs preliminary segmentation to identify and mask calcium inclusions and define fibrous cap boundaries before elasticity reconstruction. This pre-identification ensures that calcium deposits are recognized as distinct high-density regions and fibrous cap thickness can be accurately measured, while still obtaining high-quality elasticity images for the remaining soft tissue components.
Solution Approach 2:
By segmenting the plaque into distinct components including calcium deposits and fibrous cap, the method simultaneously achieves reliable component identification and accurate elasticity imaging. Each segment can be analyzed with appropriate parameters, resolving the contradiction between image quality and identification reliability.
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
Enables accurate estimation of plaque vulnerability by providing a complete morphological description and elasticity image, enhancing the prediction of plaque rupture risk by considering multiple necrotic bodies and calcium inclusions, thereby improving diagnostic precision.
Implementation Method 1
an image called an elastogram, showing internal deformations resulting from the compression of blood vessel tissue, analyzed as a function of blood pressure
Implementation Method 2
the pre-segmentation is performed considering the tissue to be an incompressible elastic solid, so as to determine the gradient of Young's modulus
Implementation Method 3
The iterative approaches use an optimization algorithm to minimize the error between measured strain and computed strain by means of a numerical model
Data Source
AI summary
The invention relates to an image processing method for estimating a risk of atheroma plaque breakage, that includes the step of receiving (40) a so-called elastogram image representing the inner deformation resulting from the compression of an analyzed blood vessel tissue based on blood pressure, the step of pre-segmenting (50) the elastogram in order to obtain a pre-segmented image including a plurality of areas, and the step of calculating an elasticity image (60) representing the elasticity of at least one region of the elastogram, the region corresponding to an area selected among the plurality of areas of the pre-segmented image, and the elasticity image enabling the user to estimate the risk of atheroma plaque breakage, wherein said method is characterized in that: the pre-segmentation is carried out while considering the tissue as being a non-compressible elastic solid in order to determine the Young modulus gradient as equal to the sum of the Lagrange multiplier gradient and of a space depending vector, and in that the pre-segmentation includes determining the outlines of each area in the plurality of areas using said space depending vector.