Physics-Informed Neural Networks for Cardiac Activation Mapping

Abstract

A critical procedure in diagnosing atrial fibrillation is the creation of electro-anatomic activation maps. Current methods generate these mappings from interpolation using a few sparse data points recorded inside the atria; they neither include prior knowledge of the underlying physics nor uncertainty of these recordings. Here we propose a physics-informed neural network for cardiac activation mapping that accounts for the underlying wave propagation dynamics and we quantify the epistemic uncertainty associated with these predictions. These uncertainty estimates not only allow us to quantify the predictive error of the neural network, but also help to reduce it by judiciously selecting new informative measurement locations via active learning. We illustrate the potential of our approach using a synthetic benchmark problem and a personalized electrophysiology model of the left atrium. We show that our new method outperforms linear interpolation and Gaussian process regression for the benchmark problem and linear interpolation at clinical densities for the left atrium. In both cases, the active learning algorithm achieves lower error levels than random allocation. Our findings open the door toward physics-based electro-anatomic mapping with the ultimate goals to reduce procedural time and improve diagnostic predictability for patients affected by atrial fibrillation. Open source code is available at https://github.com/fsahli/EikonalNet.

Biologically-informed neural networks guide mechanistic modeling from sparse experimental data

“…This methodology ensures that the neural network solution satisfies the physical laws described by F while simultaneously fitting the spatiotemporal data. Theory-informed neural networks have been demonstrated with smaller amounts of data in the presence of noise, however, they have so far only been applied to problems where the governing mechanistic PDE is known a priori [30][31][32].…”

Precision medicine in human heart modeling

“…Our findings suggest that computational modeling can identify non-local deflections to improve activation mapping and explain how and where ablation can terminate persistent atrial fibrillation ( Sahli Costabal et al 2018 ). Innovative technologies that enable real-time interactive simulations of cardiac electrophysiology ( Kaboudian et al 2019 ; Vasconcellos et al 2020 ), for example based on physics-informed neural networks for cardiac activation mapping ( Sahli Costabal et al 2020 ), are an important step to translate these computational tools into clinical practice.…”

Multiscale Modeling Meets Machine Learning: What Can We Learn?

“…Recent trends in computational physics suggest that exactly this approach [14]: to create data-efficient physics-informed learning machines [96,97]. Biomedicine has seen the first successful application of these techniques in cardiovascular flows modeling [50] or in cardiac activation mapping [109], where we already have a reasonable physical understanding of the system and can constrain the design space using the known underlying wave propagation dynamics. Another example where machine learning can immediately benefit from multiscale modeling and physics-based simulation is the generation of synthetic data [106], for example, to supplement sparse training sets.…”