Effect of the Conduction Network Structure of the Heart on Modeling of the Body Surface Potential Map and the ECG

1. Introduction

The reliability of its computations on all factors makes it difficult to develop a realistic simulator for cardiac activation. It is possible to model the body as either an inhomogeneous or homogeneous volume conductor. It is assumed that all organs, including the blood volume inside ventricles, have the same physical characteristics (permeability, conductivity) for a homogenous volume conductor. An inhomogeneous volume-conductor, on the other hand, has characteristics specific to each organ. Gulrajani and Mailloux [1] presented the idea that the generated surface potential is impacted by the body's inhomogeneity.

In actuality, every organ is made of anisotropic materials. While some tissues, like the liver and lungs, have nearly isotropic qualities, others, like the skeletal muscles, have considerable anisotropic qualities [2]. An enormous quantity of information on the makeup of each organ is needed to account for organ anisotropy. It is okay to treat the organs as an isotropic material, nevertheless, as this would make modeling easier [2]. The realistic torso shape is the most commonly used model, however some models define the volume conductor (the body) in terms of an estimated shape [3]. Either a homogeneous volume conductor, as reported in the literature [4,5,6], or the popular inhomogeneous volume conductor, as described in [7,8,9,10], are the realistic torso form models.

Ventricular conduction networks are modeled by either constructing a network based on anatomical structure and activation isochrones [14,15,16] or allocating the early activation locations based on Durrer et al. observations [11,12,13]. It was also observed that choosing the right excitation sites is a very sensitive task, since a tiny variance in conduction sites will result in a big variation in results. Creating such a network is always accomplished by trial and error. Since they employ certain experimental pacing sites (primary sources) in their models, other model groups do not model any conduction systems [17,18,19,20]. The impact of two conduction networks on the Body Surface Optional Map (BSPM) is compared in this study.

2. Methods

2.1. The Human Torso and the Human Heart Modeling

A realistic human torso (Figure 1), based on CT scans [21], has been created. It features various organs, and a human heart [22] is inserted at the right place inside the torso to replicate the electrical activity of the heart as a volume source inside a volume conductor.

2.2. The Conduction Network Modeling

Two conduction networks are analyzed (Figure 2) in order to determine how the conduction network structure affects the Body Surface Potential Map (BSPM) [23, 24]. The locations of the trabecular muscles are used to manually construct the first model (Model 1), and the Diffusion Volume quantity is used to extract the second model (Model 2). It is evident that the second model is more accurate than the first.

2.3. Activation Isochrones Modeling

For both conduction system models, the normal activation excitation propagation (QRS complex only) is implemented [25,26,27]. Both models' excitation isochrones are displayed in Figure 3 and Figure 4.

2.4. The Body Surface Potential Map (BSPM) Calculation

Figure 5 and Figure 6 show the BSPM of both conduction network models, which are computed for the excitation propagation of the normal activation [28] and are shown as frames on each 10 mSec.

3. Results

3.1. Body Surface Potential Map Validation

The BSPM generated by the Forward Model of both Conduction Systems (Model 1 and Model 2) is comparable to a reference model derived from the General Public Licensed (GUN) simulator ECGSIM [29] (Figure 7); however, the Model 2 conduction network's results are superior to Model 1 since they are significantly closer to the reference model. This suggests that even little modifications to the conduction network result in notable variations in the BSPM that is produced.

3.2. Validation of the ECG

Referring to actual measurements [30], the 12-Lead ECG electrograms (Figure 8) are likewise produced for both conduction networks (Figure 9). The results of both models are similar to those of the reference model.

4. Conclusion

This study demonstrates that the structure and accuracy of the ventricular conduction network play a critical role in the fidelity of body surface potential maps. Comparing two models, it was shown that the network derived from diffusion volume data (Model 2) produces BSPMs that more closely match reference data from ECGSIM, underscoring the importance of anatomically and functionally accurate conduction networks in cardiac modeling. These findings suggest that realistic conduction system modeling is essential for developing reliable forward simulation models in cardiac electrophysiology.