An electric model of the human heart

The implementation of physics-informed approaches to the study of the human heart is an inevitable step in analytically explaining much of the empirical evidence that has been traditionally used to diagnose and treat patients. Such approaches are not new and extensive work has been done in the application of mathematical models to problems relating to the electrical functioning of the human heart.

However, these in silico studies are conducted using geometries that have no resemblance to the organ in question, simulating only patches of tissue rather than the whole heart. In this work, we present a model of the whole human heart's electrical conduction system and a procedure to compute the 3 first leads of the standard 12-lead Electrocardiogram.

The model and the ECG procedure are then compared with clinical ECG data from a database of Norwegian endurance athletes [1]. The results are in great agreement with the ECG recorded for athlete 5 of the mentioned database. Following the benchmarking of both the model and the ECG procedure, pathologies relating to the heart's electrical conduction system are introduced to analyse the results of the respective ECG.

Some of the results of this analysis highlight shortcomings of the model, but in general, the ECG waves are in agreement with clinical data. We conclude that it is viable to create a simple model from both a geometrical and computational standpoint, while still achieving clinically relevant results.