Effect of Interelectrode Distance on ECG Potentials - Modelling Approach vs. Clinical Data

Merja Puurtinen, Jari Hyttinen, Jaakko Malmivuo

Ragnar Granit Institute, Tampere University of Technology, Tampere, Finland

Correspondence: M Puurtinen, Ragnar Granit Institute, Tampere University of Technology, P.O. Box 692, FIN-33101 Tampere, Finland.
E-mail: merja.puurtinen@tut.fi, phone +358 3 3115 2117, fax +358 3 3115 2162

1.  Introduction

The cardiac activity is commonly monitored by the standard 12-lead system. However, the advancements in wireless and portable technology suggest that the possibilities for smaller measurement devices and configurations should be surveyed. It is assumed that decreasing the IED decreases the measuring depth. Accordingly, this causes that signals arising from deeper sources, such as the heart, decrease. Nevertheless, the magnitude of the effect of reducing the IED has not been widely studied. This paper discusses a project where the effect of reducing the IED on the signal strength was modeled and tested with real ECG database.

2.  Material and Methods

The model used in this project was an axial, segmented 2D image from a male cadaver thorax, which included all visible details and 36 tissue types (Fig.1). The anatomy of the model was based on Visible Human Man data and extracted from full thorax model [Kauppinen et al., 1998]. The study utilized a bioelectric field software created at Ragnar Granit Institute and was based on Finite Difference Method (FDM), in which the model composes of cubic elements forming a resistor network.

The calculation of the sensitivity of the leads to measure hearts electric activity, lead field, was based on the theorem of reciprocity [Malmivuo and Plonsey, 1995]. A reciprocal current was applied on a pair of electrodes located on the model, and thus the resulting electric field in the heart muscle represented the leads sensitivity to measure the electric source of the heart. The calculations were conducted with electrodes corresponding axially to the standard 12-lead system precordial leads V1-V6, and with additional interposed electrodes (Fig. 1). The lead field was calculated for each electrode pair consecutively; firstly for standard leads (i.e.V1-V2) and then for reduced interelectrode distances. Overall, the calculated distances were ¼, ½ and ¾ of the standard distance between two adjacent precordial leads.

Figure 1. Image of the axial thorax FDM model (showing only fat and heart muscle tissues) and the locations of precordial electrodes and additional electrodes used in lead field calculation.

For evaluating the modeling results, clinical potential map data of 120-leads ECG acquired from 236 patients was analyzed. On the area of the precordial leads, the data provided signals recorded with ½ of the standard electrode distance. From this data, the maximum average QRS amplitude from given electrodes was calculated.

3.  Results

Figure 2. illustrates normalized results calculated with electrodes V1-V2, V5-V6, and with their interposed electrodes. The results show that reducing the IED evidently decreases the signal strength, but the degree of this decrease varies depending on the lead. The correlation factor between model and clinical data was 0.6913 when all results were included. Yet, when two outliers (leads detecting the direction of hearts’ electrical axis, V2-V4 and V2-V5) were ignored, correlation increased to 0.9169.

Figure 2. Normalized values for electric field (model data) and signal amplitude (clinical data) with different interelectrode distances in leads V1-V2 and V5-V6. Distance 1 is the original distance between V1 and V2 or V5 and V6, and the electric field and signal amplitude values are normalized with the value obtained with this original distance.

4.  Discussion

The results obtained from the model represent fairly well the clinical data. Even though the model is only a 2D sample of human chest representing one person, the results show that the model calculations give an indicative of the obtained signal level. Further study with 3D will be conducted.

Nevertheless, the model only takes into account the chests anatomy, and therefore it does not consider the direction or the exact location of hearts electric source. For this reason, only the signal strength can be estimated, not the pattern. Still, the results suggest that the model’s lead field can be used to predict the signal strength and thus to optimize the electrode distance and location.

Acknowledgements

The authors gratefully acknowledge that the clinical data was supplied by Fred Kornreich, and that this work was supported by Ragnar Granit Institute and the Academy of Finland.

References

Kauppinen P, Hyttinen J, Heinonen T, Malmivuo J. Detailed model of the thorax as a volume conductor based on the visible human man data. Journal of Medical Engineering & Technology, 22 (3): 126-133, 1998.

Malmivuo J, Plonsey R. Bioelectromagnetism: Principles and Application of Bioelectric and Biomagnetic Fields. Oxford University Press, New York, 1995.