Key result
The presented Bayesian active learning method achieved a 99.65% reduction in computational cost compared to direct MCMC sampling while maintaining high accuracy in estimating the posterior probability density function of cardiac electrophysiological model parameters.
Why the study?
Direct MCMC sampling of cardiac electrophysiological model parameters is computationally intensive, while efficient surrogate models suffer from limited accuracy.
A novel Bayesian active learning method improves the accuracy and computational efficiency of estimating parameters in 3D cardiac electrophysiological models.
Answers are grounded in this paper and cite their sources.
Enables efficient MCMC-based personalization of cardiac models; leaves open clinical validation and adoption.
Probabilistic estimation of cardiac electrophysiological model parameters serves an important step toward model personalization and uncertain quantification. The expensive computation associated with these model simulations, however, makes direct Markov Chain Monte Carlo (MCMC) sampling of the posterior probability density function (pdf) of model parameters computationally intensive. Approximated posterior pdfs resulting from replacing the simulation model with a computationally efficient surrogate, on the other hand, have seen limited accuracy. In this study, we present a Bayesian active learning method to directly approximate the posterior pdf function of cardiac model parameters, in which we intelligently select training points to query the simulation model in order to learn the posterior pdf using a small number of samples. We integrate a generative model into Bayesian active learning to allow approximating posterior pdf of high-dimensional model parameters at the resolution of the cardiac mesh. We further introduce new acquisition functions to focus the selection of training points on better approximating the shape rather than the modes of the posterior pdf of interest. We evaluated the presented method in estimating tissue excitability in a 3D cardiac electrophysiological model in a range of synthetic and real-data experiments. We demonstrated its improved accuracy in approximating the posterior pdf compared to Bayesian active learning using regular acquisition functions, and substantially reduced computational cost in comparison to existing standard or accelerated MCMC sampling.
No takes yet. Share an insight, caveat, or question.
Zaman et al. (2021) studied Myocardial infarction and ventricular tachycardia (n=9). Bayesian active learning with generative modeling and posterior-focused acquisition functions vs. Direct MCMC sampling, two-stage MCMC, and regular Bayesian active learning was evaluated on Accuracy (KL divergence) and computational cost of estimated posterior probability density function. The presented Bayesian active learning method achieved a 99.65% reduction in computational cost compared to direct MCMC sampling while maintaining high accuracy in estimating the posterior probability density function of cardiac electrophysiological model parameters.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: