Cardiac Hemodynamic Digital Twin Personalization Using Echo and PSO
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cardiac digital twin models lack personalized lumped hemodynamic models due to the large number of dependent and independent model parameters that require tuning from limited clinical data, making them computationally expensive and difficult to implement.
Innovation Solution
A method and system using echocardiogram-based approaches to personalize cardiac hemodynamic digital twins through particle swarm optimization (PSO) for optimizing model parameters such as left ventricle height, active and passive pressure components, and compliance parameters, utilizing echocardiogram (ECG) data and echo data to derive optimal values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CFD models with high resolution 3d medical images are used for personalized hemodynamic modeling, then measurement precision and model accuracy are improved, but device complexity and computational cost increase significantly
Solution Approach 1:
The patent segments the complex hemodynamic modeling process into distinct components: geometry extraction from 3D images, parameter estimation using echo data, and CFD simulation. This segmentation allows each component to be optimized independently, reducing overall system complexity while maintaining accuracy.
Solution Approach 2:
The patent transforms the modeling approach by changing parameters from direct CFD simulation of entire cardiovascular systems to a reduced-order model that estimates key hemodynamic parameters (pressure, flow, compliance) using echocardiogram data. This parameter transformation dramatically reduces computational complexity while preserving clinical accuracy.
2Device complexity
If reduced order models are used for hemodynamic functionality, then device complexity and computational cost are reduced, but measurement precision and model accuracy deteriorate
Solution Approach 1:
The patent introduces an intermediary parameter estimation step that uses echocardiogram data to derive physiological parameters (pressure components, compliance, resistance) that bridge the gap between simple reduced-order models and complex CFD simulations. This intermediary layer maintains accuracy while reducing computational burden.
Solution Approach 2:
The patent replaces complex mechanical CFD simulations with a physics-informed reduced-order model that uses analytical solutions and empirical relationships from echocardiogram data. This substitution maintains hemodynamic accuracy while eliminating the need for computationally intensive numerical simulations.
3Measurement precision
If manual tuning of model parameters is performed, then measurement precision is improved, but productivity and time efficiency worsen
Solution Approach 1:
The patent implements self-service by enabling the model to automatically estimate and optimize its own parameters using the patient's echocardiogram data. The system performs autonomous parameter calibration without requiring manual expert tuning, thereby maintaining accuracy while dramatically improving productivity and reducing time to personalization.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Existing techniques fail to propose a method for improving model parameters optimization and seamless integration of clinical data with computational model required for personalized cardiac hemodynamic model development. This disclosure relates to a system and method, which receives one or more model parameters comprising height, active pressure components and passive pressure components specific to left ventricle from cardiac hemodynamic model. The first set of values corresponding to one or more input parameters comprising end systolic diameter and end diastole diameter are received from subject specific Echo data. The second set of values corresponding to one or more model parameters are estimated using one or more input parameters. The estimated second set of values are optimized using a Particle swarm optimization to obtain one or more optimized values. Select at least a subset of optimized one or more values corresponding to one or more model parameters for personalization.