Quantifying the Uncertainty in Model Parameters Using Gaussian Process-Based Markov Chain Monte Carlo: An Application to Cardiac Electrophysiological Models

@inproceedings{Dhamala2017QuantifyingTU,
  title={Quantifying the Uncertainty in Model Parameters Using Gaussian Process-Based Markov Chain Monte Carlo: An Application to Cardiac Electrophysiological Models},
  author={J. Dhamala and John L. Sapp and Milan Horacek and Linwei Wang},
  booktitle={IPMI},
  year={2017}
}
Estimation of patient-specific model parameters is important for personalized modeling, although sparse and noisy clinical data can introduce significant uncertainty in the estimated parameter values. This importance source of uncertainty, if left unquantified, will lead to unknown variability in model outputs that hinder their reliable adoptions. Probabilistic estimation model parameters, however, remains an unresolved challenge because standard Markov Chain Monte Carlo sampling requires… 
Fast Posterior Estimation of Cardiac Electrophysiological Model Parameters via Bayesian Active Learning
TLDR
The presented Bayesian active learning method is evaluated in estimating tissue excitability in a 3D cardiac electrophysiological model in a range of synthetic and real-data experiments and demonstrates its improved accuracy in approximating the posterior pdf compared to Bayesianactive learning using regular acquisition functions.
Bayesian Active Learning for Personalization and Uncertainty Quantification in Cardiac Electrophysiological Model
TLDR
This dissertation proposes the surrogate modeling of the complex target function containing the forward model – an objective function in deterministic estimation or a posterior probability density function in probabilistic estimation – by actively selecting a set of training samples and a Bayesian update of the prior over the target function.
High-Dimensional Bayesian Optimization of Personalized Cardiac Model Parameters via an Embedded Generative Model
TLDR
A novel concept is presented that embeds a generative variational auto-encoder (VAE) into the objective function of Bayesian optimization, providing an implicit low-dimensional (LD) search space that represents the generative code of the HD spatially-varying tissue properties.
Transfer Learning From Simulations on a Reference Anatomy for ECGI in Personalized Cardiac Resynchronization Therapy
TLDR
The fast CRT pacing predictions are a step forward to a noninvasive CRT patient selection and therapy optimisation, to help clinicians in these difficult tasks.
Non-invasive Personalisation of Cardiac Electrophysiological Models from Surface Electrograms. (Personnalisation Non-invasive de Modèles Electrophysiologiques Cardiaques à Partir d'Electrogrammes Surfaciques)
TLDR
A non-invasive EP model personalisation based on a patientspecific simulated database and machine learning regressions is developed, which has comparable results with state-of-the-art ECG imaging methods and shows good correlation coefficients with the real BSPMs.

References

SHOWING 1-10 OF 19 REFERENCES
Markov chain Monte Carlo Using an Approximation
This article presents a method for generating samples from an unnormalized posterior distribution f(·) using Markov chain Monte Carlo (MCMC) in which the evaluation of f(·) is very difficult or
Uncertainty quantification in virtual surgery hemodynamics predictions for single ventricle palliation
TLDR
This study proposes a methodology for full propagation of uncertainty from clinical data to model results that enables estimation of the confidence associated with model predictions, and quantifies the statistical variability in virtual surgery predictions.
Bayesian Personalization of Brain Tumor Growth Model
TLDR
A method for conducting the Bayesian personalization of the tumor growth model parameters, based on a highly parallelized implementation of the reaction-diffusion equation, and the Gaussian Process Hamiltonian Monte Carlo, a high acceptance rate Monte Carlo technique.
Preconditioning Markov Chain Monte Carlo Simulations Using Coarse-Scale Models
TLDR
The purpose of preconditioning is to reduce the fine-scale computational cost and increase the acceptance rate in the MCMC sampling by generating Markov chains based on two-stage computations by incorporating the coarse-scale information of the problem.
MCMC can detect nonidentifiable models.
Spatially-Adaptive Multi-scale Optimization for Local Parameter Estimation: Application in Cardiac Electrophysiological Models
The estimation of local parameter values for a 3D cardiac model is important for revealing abnormal tissues with altered material properties and for building patient-specific models. Existing works
Spatially Adaptive Multi-Scale Optimization for Local Parameter Estimation in Cardiac Electrophysiology
TLDR
A novel framework is presented that, going beyond a uniform low-resolution approach, is able to obtain a higher resolution estimation of tissue properties represented by spatially non-uniform resolution and is evaluated in estimating the local tissue excitability properties of a cardiac EP model on both synthetic and real data experiments.
An Introduction to Markov Chain Monte Carlo
TLDR
The Bayesian approach to statistics is introduced, and the necessary continuous state space Markov chain theory is summarized, and two common algorithms for generating random draws from complex joint distribution are presented.
Physiological-Model-Constrained Noninvasive Reconstruction of Volumetric Myocardial Transmembrane Potentials
TLDR
A physiological-model-constrained statistical framework to reconstruct volumetric TMP dynamics inside the 3-D myocardium from noninvasive BSP recordings is presented and phantom experiments on intramural focal activities and an initial real-data study on postmyocardial infarction demonstrate the potential of the framework in reconstructing local arrhythmic details and identifying arrhythmogenic substrates inside theMyocardium.
...
...