Surrogate Model for Patient-Specific Cardiac Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computational models of human organs, such as the heart, face challenges in personalization due to high complexity, simplifying assumptions, and noisy measurements, which hinder their clinical usefulness in managing conditions like cardiomyopathy.
Innovation Solution
A method and system that utilize stochastic approaches, including polynomial chaos expansion and Bayesian inference, to estimate patient-specific model parameters and their uncertainty, using surrogate models and the mean-shift algorithm to handle noise and optimization limitations, enabling personalized cardiac modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computational models are made more complex to improve accuracy, then model precision improves, but device complexity increases
Solution Approach 1:
The patent creates a surrogate model that copies the essential behavior of the complex computational model but in a simplified form. This surrogate model is used for parameter estimation and uncertainty quantification, avoiding the need to directly manipulate the complex full model while maintaining accuracy in predicting model responses.
Solution Approach 2:
The surrogate model acts as an intermediary between the complex computational model and the parameter estimation process. It mediates by providing a computationally efficient approximation that captures the essential input-output relationships, allowing for robust parameter estimation without directly solving the complex model.
2Measurement precision
If more data is used to improve parameter estimation, then measurement precision improves, but loss of information increases due to noise
Solution Approach 1:
The patent implements a feedback mechanism through iterative optimization where the surrogate model predictions are continuously compared with measured data, and parameter estimates are updated based on the discrepancy. This feedback loop allows the system to progressively reduce the impact of noise and converge toward accurate parameter values.
Solution Approach 2:
The patent applies prior cushioning by incorporating uncertainty quantification and sensitivity analysis before final parameter estimation. By pre-assessing the impact of noise and model uncertainties, the system can weight and prioritize data sources appropriately, cushioning against the harmful effects of noisy measurements.
3Reliability
If optimization algorithms are made more robust to handle noise, then reliability improves, but computational time increases
Solution Approach 1:
The patent uses a surrogate model that copies the input-output behavior of the complex computational model but in a simplified, faster-to-evaluate form. This surrogate is used within optimization algorithms to assess objective functions and gradients, providing robustness to noise while significantly reducing the computational time required for each optimization iteration.
Solution Approach 2:
The patent applies partial action by using the surrogate model for most optimization iterations and only resorting to the full computational model when necessary. This partial use of the expensive full model maintains optimization robustness while minimizing overall computational time through the majority of iterations using the faster surrogate.
Data Source
AI summary
A method and system for estimating tissue parameters of a computational model of organ function and their uncertainty due to model assumptions, data noise and optimization limitations is disclosed. As applied to a cardiac use-case, a patient-specific anatomical heart model is generated from medical image data of a patient. A patient-specific computational heart model is generated based on the patient-specific anatomical heart model. Patient-specific parameters and corresponding uncertainty values are estimated for at least a subset of parameters of the patient-specific computational heart model. A surrogate model is estimated for a forward model of cardiac function, and the surrogate model is applied within Bayesian inference to estimate the posterior probability density function of the parameter space of the forward model. Cardiac function for the patient is simulated using the patient-specific computational heart model. The estimated parameters, their uncertainty, and the computed cardiac function are displayed to the user.