GEASI: Geodesic-based Earliest Activation Sites Identification in cardiac models

Abstract

The personalization of cardiac models is the cornerstone of patient-specific modeling. Ideally, non-invasive or minimally-invasive clinical data, such as the standard ECG or intracardiac contact recordings, could provide an insight on model parameters. Parameter selection of such models is however a challenging and potentially time-consuming task.In this work, we estimate the earliest activation sites governing the cardiac electrical activation. Specifically, we introduce GEASI (Geodesic-based Earliest Activation Sites Identification) as a novel approach to simultaneously identify their locations and times. To this end, we start from the anisotropic eikonal equation modeling cardiac electrical activation and exploit its Hamilton-Jacobi formulation to minimize a given objective functional, which in the case of GEASI is the quadratic mismatch to given activation measurements. This versatile approach can be extended for computing topological gradients to estimate the number of sites, or fitting a given ECG.We conducted various experiments in 2D and 3D for in-silico models and an in-vivo intracardiac recording collected from a patient undergoing cardiac resynchronization therapy. The results demonstrate the clinical applicability of GEASI for potential future personalized models and clinical intervention.

Source and metric estimation in the eikonal equation using optimization on a manifold

“…The reference that uses the closest approach from ours is [14], where the model is the anisotropic eikonal equation. The quadratic cost function is minimized using an over-relaxed gradient descent, and the location and timing of the sources are estimated.…”

“…60s. In order to recover two sources, we implemented the splitting method presented in [14]. It consists in starting with one single source, and at each iteration to estimate if an improvement is brought if this source is splitted in two sources.…”

“…If the gain observed by splitting is more than 5 times larger than the gain observed by moving the source, then it is decided to split the source. We refer the reader to [14] for details, we only emphasize that all the calculations are straightforward once the Log map w.r.t. the considered source is known.…”

“…The solution of the eikonal equation is only approximate but numerical simulations showed satisfactory results. The computation times are orders of magnitude faster than state-of-the-art methods [14], even though our Python implementation was not optimized.…”