High Frequency Electromagnetic Field Exposure in Paediatric and Female Patients with Implanted Cardiac Pacemaker

Abstract

This article investigates the effects of electromagnetic field (EMF) from mobile phones on human tissues and implanted medical devices. The intensity of the electric field (E) is evaluated based on simulations and measurements of various exposure scenarios. An area of interest is the case of a person with an implanted device (heart pacemaker) who may be affected by this exposure. Due to the rapid development of communication technologies and the growing awareness of the potential health risks of radio frequency (RF) EMF, the International Commission on Non-Ionizing Radiation Protection (ICNIRP) has established exposure limits within the European Union. Our study models and analyses EMF values in human tissues in an ideal environment, in a situation where a person uses a mobile phone in the DCS (Digital Cellular System) band, including the case of a person with an implanted pacemaker. Pilot simulations were verified by experimental measurements. Based on them, specific human models with the best matching results were selected for modelling other possible interactions of exogenous EMF and cardiac pacemaker in the same situations and locations.

1. Introduction

In recent years, we have seen a significant increase in the use of mobile phones in various aspects of daily life. Modern wireless communication technologies have become an integral part of our society and are practically ubiquitous—from phone applications, communication between different devices, Bluetooth connections in vehicles, Internet systems, to solutions in industrial applications. These technologies operate based on the principle of an EMF within a radio frequency (RF) [1,2,3,4,5].

The interaction of EMF with biological materials can be studied at two different levels: the macroscopic level, where we observe objects and the whole body, and the microscopic level, where cells, membranes, and molecules are the subject of study. However, the interaction phenomena at these two levels cannot be considered independently. It is necessary to consider the distribution of energy that occurs in an object located in an EMF. The macroscopic level of interaction provides information about the phenomenon of energy penetration and dissipation. At the microscopic level, interaction mechanisms can be studied at smaller scales.

The effect of the interaction of electromagnetic (EM) waves with biological tissues can be considered as the result of three phenomena: the penetration of EM waves into the living system and their propagation in it, the primary interaction of the waves with biological tissues, and possible secondary effects caused by primary interactions. The term “interaction” emphasizes the fact that the results do not depend only on the action of the field but are also influenced by the response of the living system.

EM waves propagate in tissues at a reduced speed and are refracted, diffracted, and reflected when encountering inhomogeneities. The specific electrical properties of each tissue govern the rate of reduction, refraction, and diffraction. These properties, as well as the geometry of the inhomogeneities, determine the proportion of energy absorbed by tissues. The main parameters that describe EM waves are the frequency of oscillation, the amplitude of the electric or magnetic field, and the phase angle, which defines the instantaneous state of the oscillation.

It is very difficult to fully characterize the propagation of EM fields in the human body, bearing in mind the complexity and inhomogeneous nature of biological tissues. However, with the advent of computers, it is now possible to perform highly accurate dosimetry assessments for the human body or body parts. The EM wave includes both a changing electric field and a changing magnetic field, the propagation of which is described by the differential form of complex time-harmonic Maxwell equations [6].

The principles of diffraction, scattering, wave penetration, and absorption are applied here. However, it is very challenging to apply these interactions to biological systems due to their extreme complexity and their different levels of organization in living organisms, as well as the wide range of dielectric properties of biological tissues. The human body is a complex environment composed of aqueous electrolytes, colloids, and cells, where tissues represent their suspension and composition [7].

From the perspective of dielectric properties, biological tissues exhibit the characteristics of layered lossy dielectrics. They are defined by their conductivity (σ) in Siemens per meter (S·m⁻¹) and complex permittivity ($\widehat{\epsilon }$) (-). These properties vary based on factors such as field frequency, tissue temperature, and water content. These factors influence how the EMF interacts with the biological environment. Generally, biological tissues are considered to be isotropic, linear, and homogeneous lossy dielectrics. Unlike ideal dielectrics, lossy dielectrics allow both displacement current and conduction current to flow through their volume under time-varying field exposures in a linear environment.

$$\mathbf{J}=\sigma \mathbf{E}+\widehat{\epsilon }{\displaystyle \frac{\partial \mathbf{E}}{\partial \mathrm{t}}},$$

(1)

where J (A·m⁻²) is current density vector, σ (S·m⁻¹) is conductivity, E (V·m⁻¹) is electric field strength, and $\widehat{\epsilon }$ (-) is complex permittivity. The interaction of the biological environment with the EMF can be described by the complex permittivity in the following form:

$$\widehat{\epsilon }\left(\omega \right)={\epsilon }^{\prime }-j{\epsilon }^{″}={\epsilon }^{\prime }-j{\displaystyle \frac{\sigma }{\omega {\epsilon }_0}}={\epsilon }^{\prime }\left(1-j\mathrm{t}\mathrm{a}\mathrm{n}\delta \right)={\epsilon }^{\prime }\left(1-j{\displaystyle \frac{{\epsilon }^{″}}{{\epsilon }^{\prime }}}\right),$$

(2)

where ω (rad·s ⁻¹) is the angular velocity, ${\epsilon }_0$ (F·m⁻¹) is the vacuum permittivity, ε′ is the real component of the complex permittivity (it is the relative permittivity of the dielectric ε_r, expressing the amount of energy stored in the unit volume of the dielectric during exposure to the field), and ${\epsilon }^{″}$ is an imaginary component representing energy losses in the dielectric. By the value of the loss factor (tan δ), which expresses the ratio of the imaginary and real components of the complex permittivity, we classify high-loss dielectric (tan δ >> 1) and low-loss dielectric (tan δ << 1) [8,9].

When studying EM waves in a biological environment, we are considering a linear, homogeneous, isotropic, and lossy environment. By applying the rotation vector operator to the first and second Maxwell’s equations and then using the third and fourth Maxwell’s equations, we can derive the Helmholtz wave equations outside the source.

$${\nabla }^2\mathbf{E}-\mu \sigma {\displaystyle \frac{\partial \mathbf{E}}{\partial \mathrm{t}}}-\mu \epsilon {\displaystyle \frac{{\partial }^2\mathbf{E}}{\partial t^2}}=0,$$

(3)

$${\nabla }^2\mathbf{H}-\mu \sigma {\displaystyle \frac{\partial \mathbf{H}}{\partial \mathrm{t}}}-\mu \epsilon {\displaystyle \frac{{\partial }^2\mathbf{H}}{\partial t^2}}=0,$$

(4)

where E (V·m⁻¹) is electric field strength, H (A·m⁻¹) is magnetic field intensity, σ (S·m⁻¹) is conductivity, µ (H·m⁻¹) is permeability, and ε (F·m⁻¹) is permittivity.

When dealing with field vectors that exhibit harmonic behaviour over time, we use the complexor form to represent their spatial and temporal dependencies. In this context, the quantities E_m and H_m denote the spatially dependent complex amplitudes of the complexors E and H. Wave equations take the form of their partial derivatives with respect to time, divided by the time factor e^jωt, after substituting the complexor expression.

$${\nabla }^2{\mathbf{E}}_{\mathrm{m}}-j\omega \mu \sigma {\mathbf{E}}_{\mathrm{m}}+{\omega }^2\mu \epsilon {\mathbf{E}}_{\mathrm{m}}={\nabla }^2{\mathbf{E}}_{\mathrm{m}}-\mathrm{j}\omega \mu (\sigma +\mathrm{j}\omega \epsilon ){\mathbf{E}}_{\mathrm{m}}={\nabla }^2{\mathbf{E}}_{\mathrm{m}}+{\mathbf{k}}^2{\mathbf{E}}_{\mathrm{m}}=0,$$

(5)

$${\nabla }^2{\mathbf{H}}_{\mathrm{m}}-j\omega \mu \sigma {\mathbf{H}}_{\mathrm{m}}+{\omega }^2\mu \epsilon {\mathbf{H}}_{\mathrm{m}}={\nabla }^2{\mathbf{H}}_{\mathrm{m}}-\mathrm{j}\omega \mu (\sigma +\mathrm{j}\omega \epsilon ){\mathbf{H}}_{\mathrm{m}}={\nabla }^2{\mathbf{H}}_{\mathrm{m}}+{\mathbf{k}}^2{\mathbf{H}}_{\mathrm{m}}=0,$$

(6)

where the constant k is known as the wave vector and describes the propagation of a harmonic wave in a dissipative medium.

$$\mathbf{k}=\omega \sqrt{\mu \widehat{\epsilon }}=\sqrt{-\mathrm{j}\omega \mu \left(\sigma +\mathrm{j}\omega \epsilon \right)}=\omega \sqrt{\epsilon \mu }\sqrt{1-j{\displaystyle \frac{\sigma }{\omega \epsilon }}}=\alpha -\mathrm{j}\beta ,$$

(7)

the real component α (rad·s⁻¹) represents the relative phase shift of the EM wave, while the imaginary component β (m⁻¹) represents the relative attenuation of the EM wave in the environment. We can conclude that the specific phase shift and attenuation of an EM 2021wave are dependent on the field frequency and the magnetic and conductivity parameters of the environment through which the wave propagates. By utilizing the known wave vector value, it is possible to calculate the characteristic impedance of the environment, which is defined as the ratio of the electric and magnetic field intensity phasors, expressed as

$$\widehat{Z}={\displaystyle \frac{{\mathbf{E}}_{\mathrm{m}}}{{\mathbf{H}}_{\mathrm{m}}}}={\displaystyle \frac{\omega \mu }{\mathbf{k}}}={\displaystyle \frac{\omega \mu }{\sqrt{-\mathrm{j}\omega \mu \left(\sigma +\mathrm{j}\omega \epsilon \right)}}}=\sqrt{{\displaystyle \frac{\mathrm{j}\omega \mu }{\sigma +\mathrm{j}\omega \epsilon }}.}$$

(8)

The ratio between the amplitudes of the electric and magnetic intensity is determined by the absolute value of the characteristic impedance, while the phase difference between the electric and magnetic components of the wave is determined by its argument. When a wave propagates in a medium with loss (like biological tissues), the amplitudes of its individual components decrease exponentially with an increasing attenuation coefficient β. To quantify that decrease, we use the effective depth of penetration, which is the distance at which the electric and magnetic intensity drop to e⁻¹ of their original value upon entering the medium [10,11].

$$\delta ={\displaystyle \frac{1}{\beta }}={\displaystyle \frac{1}{\sqrt{\frac{{\omega }^2\epsilon \mu }{2}}\sqrt{\sqrt{1+{\left(\frac{\sigma }{\omega \epsilon }\right)}^2}-1}}}.$$

(9)

Tissue is a heterogeneous material composed of water, dissolved organic molecules, macromolecules, ions, and insoluble substances. These components are highly organized into cellular and subcellular structures, forming macroscopic elements and both soft and hard tissues. The presence of ions plays a crucial role in interacting with the electric field, enabling ionic conductivity and polarization effects. The dielectric spectrum of biological tissue exhibits three main dispersion regions, known as alfa, beta, and gamma dispersions. Very high permittivity values characterize the low frequency alfa dispersion and can be at least partially attributed to the diffusion effects of counterions. beta dispersion occurs at medium frequencies and is primarily due to the capacitive charging of cell membranes and membranes of bound intracellular bodies (interfacial polarization). gamma dispersion is caused by the dipolar polarization of tissue water. At frequencies above a few hundred megahertz, where the tissue water response becomes the dominant mechanism, these values can be very high. However, tissues and other biological materials may exhibit more than the three principal dispersions. The parameters of dielectric properties for all biological tissues and their frequency-dependent variations are detailed in research studies, such as those based on Gabriel’s dispersion relations [6].

The situation is even more difficult in the case of the presence of an implanted device in the human body e.g., devices for supporting cardiac activity. Cardio stimulation technology today encompasses a wide range of implantable medical devices that are increasingly used worldwide. According to valid national and European legislation, cardio stimulation technology is classified as active implantable medical devices (AIMD), which means they are subject to the strictest requirements in terms of reliability and safety.

A modern cardio stimulation system is a medical device consisting of a device and one to three electrodes, depending on the patient’s type of heart disorder. This implantable system can be utilized in several situations. For instance, in treating a slow heart rhythm, one electrode is placed in the atrium or the ventricle. If the system senses no heart activity, the pacemaker (PM) emits a stimulation pulse of defined parameters. In cases of heart blocks of all degrees, one electrode detects the heart’s activity in the atrium, while the other provides delayed ventricular stimulation. Recently, it has become possible to address ventricular desynchrony caused by structural changes (heart failure) by introducing three epicardial electrodes to the left ventricle. An implanted cardioverter-defibrillator (ICD) is used when there is a risk of danger to the patient due to rapid heart rhythm (tachycardia). In this case, we can introduce only one electrode into the right ventricle, or two into the atrium and the other into the ventricle. The three-electrode system for resynchronization therapy of heart failure can also be equipped with a defibrillator (CRT-D). These devices contain sensors for adapting stimulation speed (frequency) according to the patient’s needs [12,13].

The widespread use of mobile phones in modern society has raised concerns about the potential health risks associated with the radiation emitted by these devices. Many studies have been conducted over the past two decades to assess whether cell phones pose a health risk. Excessive radiation from wireless devices has the potential to affect the human body’s biological system [14,15].

In the material environment, EMF propagate at a finite speed, which is the maximum speed of light in a vacuum, in the form of waves. Therefore, due to the corpuscular-wave dualism, this propagation can also be described as a flow of particles. In this case, we are referring to radiation. Particularly, because of EM waves, high frequency/radio EMF can penetrate and spread into the human body [16,17,18].

Even though there has been extensive research on the issue of using mobile phones in a shielded environment, most of this research has been carried out either by numerical calculations or by a theoretical approach; as such, we have the opportunity to enrich the issue with experimental measurements [19].

This study aims to create a series of simulations in the CST Studio Suite software v.2021.05 for different models of the human body, namely for a single-layer, multi-layer, and complex model of an adult woman and a child, as well as to compare the obtained results of the EMF quantities. Most previous studies have predominantly focused on male body models, leaving a significant gap in research dedicated to female and paediatric models. However, women and children are exposed to the same environmental conditions as men, making it crucial to understand how electromagnetic fields (EMF) might interact differently with their bodies. The physiological differences between males, females, and children—such as variations in tissue composition, body size, and developmental stages—can lead to different levels of EMF absorption and distinct effects.

To obtain a holistic picture and a relevant assessment of the results of the simulations, let us then carry out a series of experimental measurements for selected pilot simulations of a normal person and a person with a pacemaker in free space. In our research, we also consider the model of a child with a pacemaker, since this intervention is common in paediatric patients with bradycardia or atrioventricular block. These conditions often require the implantation of a pacemaker to ensure an adequate heart rhythm, which emphasizes the study of the effect of EMF on implanted devices for the child organism [20].

Experimental measurements will subsequently provide us with realistic data on EM interactions in simulated conditions, which will create an opportunity to compare and verify the accuracy of our models and simulations in an environment that is close to real conditions. With this integrated approach, we can obtain a comprehensive and reliable view of the impact of EM radiation in an ideal and shielded space.

2. CST Studio Suite Software Simulations

CST Studio Suite^® is a high-performance software package for 3D analysis, design, and optimization of EM components and systems. We specifically used the Finite Integration Technique (FIT), which operates on hexahedral meshes.

Within the CST Studio Suite software, we implemented a pilot series of simulations that included the creation of various EMF models and sources. In this simulation program, we first designed and optimized the antenna that served as the source of the EMF. This antenna has been modelled in detail to faithfully simulate the conditions of real EMF sources, including its frequency characteristics and field distribution.

Subsequently, we created different types of human body models within the CST Studio Suite, which we used in simulations to investigate EMF interactions with biological tissues. We started with a simple homogeneous model that consisted of a single layer of tissue. We gradually created more complex multi-layered models that included detailed anatomical structures. These models have been designed to best represent the various parts of the human body, including the head, torso, neck, and limbs, with an emphasis on the accuracy of tissue properties.

The series of pilot simulations created in this way gave us a comprehensive view of the distribution and intensity of the EMF in different scenarios, which was crucial for further experimental verification and applications in real conditions.

2.1. Source of High Frequency EMF

As a source of high frequency EMF for the simulations in this work, we used a Planar Inverted-F Antenna (PIFA), whose shape resembles an inverted letter F. It is a popular choice for portable wireless devices due to its small size, low price, and many other features [21,22,23]. It is often used in mobile technology, mainly in the Digital Cellular System (DCS) 1800 MHz band, used for GSM and LTE mobile networks. The frequency range usually includes 1710–1880 MHz, where 1710–1785 MHz represents the uplink (transmission from the mobile device to the base station) and 1805–1880 MHz for the downlink (transmission from the base station to the mobile device) [24]. These portable PIFA devices have dual polarization, which gives them the advantage of placing the antenna in any position. The structure consists of a ground plate, a patch, a dielectric, a shorting wall, and a feeding point. The patch is made of copper and is placed on a thin dielectric substrate that is installed above the ground plate. The PIFA antenna is powered by a coaxial cable and uses the inner wire of the coaxial connector, which is extended through the dielectric and attached to the radiating area. The outer shield cover is connected to the conductive plate (Figure 1). This power supply method has low parasitic radiation and is easy to implement [25,26].

PIFA antennas can be easily tuned by changing the length and position of the vertical element, which allows for precise adaptation of the antenna to the desired frequency band. The created single-band PIFA antenna works at a frequency of 1800 MHz, while parameter S11 represents the return loss of the device, which in our case for 1800 MHz is around −17dB, as can be seen in the Figure 2. The S11 parameter represents the energy reflected from the antenna and is therefore known as the reflection coefficient or return loss [27].

2.2. Human Body Models Used in Simulations

Overall, we implemented a three-level approach to EMF tissue exposure modelling. In the first step, we created a simple model (SM) of the human body, which consisted of only one homogeneous material used for the overall construction of the body Figure 3a. This model served as the basic reference point for our simulations. Subsequently, we moved on to create a more complex model that took into account the multilayer structure of tissues such as skin, subcutaneous fat, muscle, bone, and general tissue (Figure 3b). This part aimed to investigate how the effective value of the EM field intensity changes depending on the different tissue layers in the model.

Subsequently, we focused on creating the most complex models of the human body, which included all organs, tissues, and structures. We obtained these models from the CST Voxel Family library, which is designed specifically for simulation programs such as CST Studio Suite. Within the complex models, we used the adult woman model “LAURA” and the child model “CHILD” Figure 4a. The model “LAURA” (Figure 4b) is created according to the anatomy of a 43-year-old woman with standard physical parameters, namely height 163 cm and weight 51 kg. The “CHILD” model is based on the anatomical data of a 7-year-old girl with a height of 115 cm and a weight of 21.7 kg [28].

Their detailed anatomy and realistic representation of the human body allowed us to simulate the interactions between EM radiation and different parts of the human body with high accuracy and credibility. We performed simulations for various distances of the human model from the source of EM radiation, specifically at distances of 5 mm, 15 mm, 50 mm, and 300 mm. These distances represent realistic scenarios that allow evaluation of the impact of EM radiation on tissues in different situations.

2.3. Model of Implantable Device—CPM

For the modelling of the implanted device, we were inspired by the design of the Medtronic Sigma SSR306 cardiac pacemaker (CPM). The created model is composed of a head, in the design of which we preferred the material polyurethane A80 for its suitable properties. The interior of the pacemaker’s body is empty (formed by air), while its casing is constructed of Ti6Al4V titanium alloy, ensuring optimal mechanical strength and biocompatibility. A graphic representation of the model together with the parameters used for construction can be found in the attached image (Figure 5).

After model finalisation, we integrated a cardiac pacemaker into the multi-layer model of a person and complex models of a child and an adult woman. The location of the pacemaker is shown in Figure 6, where it is possible to visualize its exact position within each model used.

2.4. Simulations Results without CPM

The individual simulations 1 consisted of a designed human model, in which a source of EM radiation was strategically placed. This radiation source was represented by a PIFA antenna that was carefully configured to provide consistent and measurable EMF outputs. The antenna power was fixed at 0.5 W, which ensured a consistent radiation source for all simulations. This performance was chosen to be high enough to induce measurable responses in models, but at the same time safe and representative for common applications in practice.

The simulations were performed in an ideal space. Within the CST Studio Suite software, we set the boundary conditions to open to simulate an infinite space without reflections that could affect the propagation of EM waves. The background properties were set to normal, meaning that the simulations took place in an environment with standard EM properties. We performed simulations for four distances between the radiation source and the human model: 5 mm, 15 mm, 50 mm, and 300 mm. These distances have been carefully chosen to cover a wide range of real-world situations, from very close contact to relatively long distances. Each of these distances provided us with valuable data on how the intensity and distribution of the EMF change its properties depending on the distance from the source. The highest effective electric field intensity values of simulations 1 for the simple model (SM 1), the multilayered model (MM 1), the model “CHILD” (Child 1) and the model “LAURA” (Laura 1) without CPM are placed in the

Table 1. The highest effective electric field intensity values for each model in the simulations at the specified distances without CPM (simulations 1).

Distance	5 mm	15 mm	50 mm	300 mm
E [V/m] by CST—SM 1	270.43	95.18	40.85	3.01
E [V/m] by CST—MM 1	301.43	100.57	41.89	3.05
E [V/m] by CST—Child 1	361.03	106.01	43.12	3.26
E [V/m] by CST—Laura 1	345.28	105.17	42.51	3.12

.

2.5. Simulations Results with CPM

The simulations for the models with the implanted device (simulations 2) were performed under the same simulation conditions as the previous tests, which allowed us to ensure the consistency and comparability of the results. We chose three different models for these simulations: a multi-layered model of a person, a complex model of a child, and a complex model of a woman. The simple human model was omitted in this series of simulations, as in previous results it showed the greatest deviations in the resulting values compared to complex models.

Simulations for each of the selected models were created for four different distances between the source of EM radiation and the model: 5 mm, 15 mm, 50 mm, and 300 mm. These distances were carefully chosen to cover a wide range of real-world situations and to allow analysis of the effect of distance on the distribution and intensity of the EMF in the presence of an implanted device.

Through these simulations, we have gained valuable knowledge that contributes to a better understanding and prediction of the behaviour of the EMF in the presence of implanted devices, which is important for improving the safety and efficiency of such devices in real conditions. The highest effective electric field intensity values of simulations 2 for the simple model (SM 2), the model “CHILD” (Child 2) and the model “LAURA” (Laura 2) with CPM are placed in the

Table 2. The highest effective electric field intensity values for each model in the simulations at the specified distances with CPM (simulations 2).

Distance	5 mm	15 mm	50 mm	300 mm
E [V/m] by CST—SM 2	408.28	159.63	103.55	54.96
E [V/m] by CST—Child 2	453.77	181.05	119.84	65.34
E [V/m] by CST—Laura 2	449.17	177.97	116.38	62.81

.

3. Experimental Measurements

These measurements were designed to imitate the conditions used in the simulations as closely as possible and to allow direct comparison and verification of our numerical models. When preparing the experimental measurements, we focused on the replication of all relevant parameters, including the geometry of the model, the distance between the source of the EMF and the model, as well as the power settings of the antenna. The experimental setup was carefully calibrated to minimize possible sources of error and ensure the greatest possible accuracy of the measured data. Overall, the experimental measurements served as a critical step in the process of verification and validation of our simulation results, thereby ensuring higher credibility and accuracy of our research conclusions. These experimental data have significantly contributed to a deeper understanding of EMF interactions with biological tissues and implanted devices, which is essential for improving the safety and effectiveness of such devices in clinical practice.

3.1. Experimental Set-Up

We conducted the experimental measurements in an attenuation chamber, which offers an optimal environment for precise and reproducible measurements under controlled conditions. This is essential for obtaining dependable results. The measurements will provide important information about exposure to EM radiation, which is crucial for assessing potential health impacts and developing safer technologies and equipment that minimize EMF exposure. By finding the effective value of the electrical intensity on the surface of the human phantom, we aim to gain a deeper understanding of how EMF interacts with the human body. To perform the measurement, we utilized the ANRITSU high-frequency signal generator and connected it to the broadband Horn antenna DRH 18-E using coaxial cables and SMA connectors. The generator was set to a frequency of 1.8 GHz with a maximum output power of 20 dBm. We then prepared a measuring table on which a phantom and ANRITSU spectrum analyser were placed. The VERT900 Vertical Antenna receiving probe was connected to the spectrum analyser using a connector with an attenuation element. The synthesized signal generators in the MG369XB series are high-resolution signal sources that are microprocessor-controlled and capable of phase-locking. They can produce discrete CW frequencies as well as wideband (full range) and narrowband step bands within the frequency range of 2 GHz to 67 GHz. Additionally, there are options available to expand the lower limit of the frequency range to 0.1 Hz. The signal generator’s functions can be controlled locally via the front panel or remotely (excluding on/off) through the IEEE-488 general purpose bus (GPIB).

The Broadband Horn antenna DRH 18-E is a high-precision antenna designed for a wide range of applications. It can be used to evaluate exposure to EM radiation in the microwave band, accurately measure energy ratios of microwave radiation to assess population radiation safety, and measure parameters of all types of EMF where high measurement accuracy across a wide frequency range is required (1 MHz to 19 GHz) (Figure 7). Technical data of the antennae are listed in

Table 3. The technical data of the Horn antenna DRH 18-E.

parameter	technical data
frequency range	1 GHz ÷ 19 GHz
dimensions	24.3 cm × 20 cm × 15.4 cm
connector type	SMA (50 Ω)
frequency-dependent characteristics	absolute gain Gi [dBi]
directional characteristic shape	directional SCH, which narrows with increasing frequency

.

The Vertical Antenna functions as a receiving probe. It is a dual-band vertical antenna specifically designed for use in the 824–960 MHz and 1710–1990 MHz frequency bands, making it well-suited for GSM, 3G, and some 4G LTE applications. With a gain of 3 dBi, this antenna provides reliable coverage in both bands, making it suitable for various wireless communication needs, including mobile networks and fixed wireless applications.

The Anritsu Spectrum Master is designed to monitor, measure, and analyse signals VERT900 in the frequency range from 100 kHz to 3 GHz. Typical measurements include in-band interference, transmission spectrum analysis, antenna isolation, and cell area interference. It offers a range of marker functions, such as peak, mid, and delta functions, for faster and more comprehensive measurement of displayed signals.

The overall measuring set is presented in the block diagram shown in Figure 8.

A human body phantom with the following characteristics and composition is used for measurements (see Figure 9). The phantom consists of a human torso bust filled with physiological NaCl solution, simulating the conductive internal environment of the human body. This solution mimics the fluids in the human body, allowing realistic modelling of EM interactions. The surface of the phantom is covered with pigskin, and a pacemaker head is placed on the surface of the bust under the skin. The receiving probe is placed on the surface of the phantom’s skin in position above the pacemakers both without and with pacemaker measurements. This sequential probe placement ensures accurate recording of EM interactions between the pacemaker, skin, and probe, which is crucial for understanding and evaluating the impact of these devices on the human body.

To simulate a human body, we fill a human bust with saline to create a conductive internal environment. The head of the pacemaker is then placed in the chosen position, and a receiving probe is placed on the surface of the bust, ensuring proper positioning and connection. Before starting the measurement, we set the radiating device to 0.5 W and ensure the frequency is set to 1.8 GHz. We conduct the first measurement at a distance of 5 mm from the antenna and repeat the measurement 10 times for accuracy. Subsequently, we increase the distance from the radiating antenna to 15 mm, 50 mm, and 300 mm, repeating the measurement 10 times at each distance. This procedure allows us to verify measurement consistency and evaluate how the electric intensity vector changes with distance from the source. We measure the electric intensity vector values on the surface and under the skin layer of the human phantom during each 10-s interval.

During the experiment we measure without the pacemaker in the position of its placement on the surface of the created phantom and in the same position with a CPM under the skin of the phantom. For all experimental measurements, we follow the specific procedure, enabling a direct comparison of the effect of the implanted device on the EM interactions in the tissue with the previous results.

We are conducting measurements at a frequency of 1.8 GHz, which is the standard frequency for the LTE (Long Term Evolution) system. This frequency is important for realistic communication systems, ensuring the relevance of our measurements. To ensure accuracy, we set the total power of the radiating device to 0.5 W, considering potential power losses due to attenuation at the connectors. Each measurement is taken over a 10-s interval to gather detailed EMF data.

After completing the measurements, we analyse the data to determine the effective value of the electric intensity vector on the surface and under the skin of the human phantom. Our goal is to evaluate the impact of the radiating device on EMF in this environment. Understanding the results of EM measurements involves converting the units of the measured signal to evaluate the intensity of the EMF accurately. This process includes several steps that transform the results from one form to another.

To obtain the actual power level (dBm_r) from the measured value (dBm_m), we need to adjust for attenuation. We use a 20 dB attenuator to ensure that the readings reflect real conditions.

Then we convert the corrected real power value in dBm_r to the equivalent in microvolt decibels (dBμV). The system impedance in ohms is set to 50 Ω as for most RF systems by default. This conversion is based on converting 1 mW to 1 μV in a 50 Ω system, which defaults to 107 dBm = 1 μV. The result, in dBμV, represents a voltage value that enables comparison of signals based on their voltage levels rather than power levels. This can be more useful for certain analyses and applications, such as compatibility and interference studies in EMF. Subsequently, dBμV is converted to dBμV/m, providing a measurement of electric field strength in units of decibel microvolts per meter. This unit is essential for quantifying the field strength in a specific space and considers the distribution of the field around the source.

$${\displaystyle \frac{\mathrm{d}\mathrm{B}\mathsf{\mu }\mathrm{V}}{\mathrm{m}}}=\mathrm{d}\mathrm{B}\mathsf{\mu }\mathrm{V}+\mathrm{A}\mathrm{F}$$

(10)

where dBμV is a unit of absolute level relative to a voltage of 1 µV (microvolt) and AF represents the antenna factor. The antenna factor is important for converting the voltage value into the field strength value. It indicates how efficiently an antenna converts an EMF into an electrical signal (voltage) at a specific frequency. This conversion process allows the measured voltage to be transformed into a more useful metric that shows the amount of EMF strength present at a particular point relative to the antenna. The resulting dBμV/m is used for more accurate and relevant comparisons of exposure to EM radiation in different environments or under different conditions.

The last step is to convert dBμV/m to V/m (volts per meter), which represents the absolute value of the electric field strength. This conversion is crucial for the final exposure and safety assessment, as volts per meter is the standard unit for expressing EMF strength in international standards and guidelines.

$${\displaystyle \frac{\mathrm{V}}{\mathrm{m}}}={10}^{\left({\displaystyle \frac{\frac{\mathrm{d}\mathrm{B}\mathsf{\mu }\mathrm{V}}{\mathrm{m}}-120}{20}}\right)}$$

(11)

In the previous conversion step, dBμV/m represents the electric field strength value in decibel microvolts per meter. The value 120 is used as the reference value to calibrate the scale between dBμV/m and V/m. This value serves as the conversion factor that transforms dBμV/m into a linear V/m scale. This conversion step yields the resulting electric field strengths in units that are directly relevant to engineering specifications and health regulations. V/m is directly applicable for evaluating whether EMF levels exceed the limit values established by health and safety standards, which is essential for protecting human health from potential adverse effects of high levels of EM radiation.

3.2. Experimental Results without CPM

The values obtained from the experimental measurements are directly used for further processing. After making the initial preparations, we began the actual measurement following the established procedures described in the previous section. Prior to measuring the values, we accurately determined the distance between the antenna and the chest phantom.

Once the measuring set was connected to the electrical network, we gradually obtained the power level values for each predetermined distance. We repeated this procedure ten times for each distance to ensure an adequate number of measurements. Upon completion of all measurements, we calculated statistical parameters such as the arithmetic mean, median, maximum value, and standard deviation for each set of values. These statistical data provide a more detailed understanding of the characteristics of the measurements and their variability at individual distances from the radiation source.

The highest effective electric field intensity values of measurements 1 without CPM are placed in the

Table 4. The highest effective electric field intensity values for the measurements at the specified distances without CPM (measurements 1).

Distance	5 mm	15 mm	50 mm	300 mm
E [V/m] by measurements 1	126.33	90.47	39.49	3.1

.

3.3. Experimental Measurements with CPM

In the following subsection, we performed experimental measurements in the attenuation chamber, placing an implanted device inside the skin of the phantom—specifically, a pacemaker. As in the previous measurements, we kept the same procedure, which allows a direct comparison of the effect of the implanted device on the EM interactions in the tissue with the previous results.

During the experimental measurements, we recorded detailed results of individual measurements, while we focused on obtaining the most accurate data possible for subsequent analysis. Each measurement set was subjected to a thorough statistical analysis, which included the calculation of basic statistical parameters such as arithmetic mean, median, maximum value, and standard deviation.

To ensure the consistency and correctness of the data, we subsequently recalculated the maximum values from individual measurements according to the procedure described in detail in the previous chapter. This procedure involved the application of correction factors and adjustments based on the calibration measurements to eliminate any systematic errors and ensure the accuracy of the final results.

The resulting maximum values were recorded in a clear table (

Table 5. The highest effective electric field intensity values for the measurements at the specified distances with CPM (measurements 2).

Distance	5 mm	15 mm	50 mm	300 mm
E [V/m] by measurements 2	166.53	127.79	105.07	53.27

). This table provides a comprehensive view of the maximum values of EMF exposure for different distances and is a key tool for further interpretation and analysis of the results.

3.4. Comparison of Simulations and Measurements Results without CPM

The resulting values for simulations 1 and measurements 1 without CPM are summarized in graphic display in Figure 10.

The composition of the models at a very close distance (5 mm) from the source of EM radiation plays a key role. Neither the phantom used in the experimental measurements, nor the simple human model (SM) used in the simulations contain bones, and the differences between their E values are the smallest compared to the other simulated models. Based on these observations, we assume that bones can be an important factor influencing the distribution of EMF in the human body, from which could reflect some of the radiation, creating local maxima on the body surface.

Comparing the multi-layer model (MM) with the simple model (SM) in terms of their structure, we find that the MM is composed of five layers and contains bone structure in the rib area. This difference in structure may account for the differences in peak E values between MM and complex chest models. While MM represents the internal organs with a layer of general tissue, complex chest models contain more detailed anatomical structures, including all the bones of the trunk, head, and upper limbs, as well as blood in the circulatory system and air in the respiratory system.

Another important factor that we take into account at a very close distance is the issue of the very near field, which is still little explored and thus unpredictable. There is an adaptation of the antenna and the biological tissue, which is close to the conductive connection. The formulae used to calculate the power level are primarily intended for the far field, which can lead to deviations at small distances. This effect is also seen at larger distances, such as 15 mm, 50 mm, and 300 mm, where the deviations between measurements and simulations are much smaller. With a 15 mm separation, the SM model shows variances of 4.95% and the Child model of 14.66%. The variations for the SM and Child models at a 50 mm distance are 3.33% and 8.41%, respectively. The variances at a distance of 300 mm are 2.91% less for the SM and 4.91% for the Child model are.

These outcomes indicate that the detailed structure of the models, including the presence of bones and other anatomical details, has a significant effect on the distribution and intensity of EMF in the body. This effect is most pronounced at very close distances, where local maxima of the field strength may occur due to reflections from bone and other structures. In the case of larger distances, these effects are less pronounced, which is also reflected in smaller deviations between experimental measurements and simulations.

3.5. Comparison of Simulations and Measurements Results with CPM

The resulting values for simulations 2 and measurements 2 with CPM are summarized in graphic display in Figure 11.

Observed differences between the resulting values from measurements and simulations at a distance of 5 mm reach more than 50%. We assume that such a significant deviation can be caused by several factors, which are listed below. One of the main reasons may be the complexity and number of tissue structures included in the simulated models compared to the phantom used for the experimental measurements. The simulated models contain more detailed representations of human tissues, including bones, muscles, and internal organs, which can affect the EMF interaction and the resulting E values.

Another significant difference is the actual pacemaker model used in the simulations. To reduce the computational and time requirements, we used a simplified model of a pacemaker, which is hollow or consists only of air, in the simulations.

The third factor is the distance itself and the characteristics of the near field (up to a distance of several millimetres). The formulas used to convert power levels (E values) are often defined for the far field, which can lead to inaccuracies at minimal distances.

When comparing the results for larger distances, the deviations between measurements and simulations are significantly reduced: at a distance of 15 mm, the deviations are reduced to less than 30%; at a distance of 50 mm, the deviations are less than 12.33%; and at a distance of 300 mm, the deviations are less than 18.48%.

The highest deviations for all distances were recorded for the complex model of the child’s chest (CHILD). This phenomenon may be due to the different physiology of the tissue structures of children compared to adults, as well as smaller body dimensions, which affect the distribution and interaction of EMF [29,30].

The results of our analysis show that the composition of the models, the complexity of the pacemaker, and the near-field characteristics are key factors influencing the differences between experimental measurements and simulations. For more accurate results in simulations, it is necessary to take these factors into account and optimize models and calculation methods, especially at minimal distances. The detected deviations thus raise new questions and emphasize the need for further research in the area of the near field and its impact on implanted medical devices.

4. Discussion and Conclusions

From the previous evaluations of the subsection, we believe that significant deviations between experimental measurements and simulations are caused by the following factors:

Most simulated models include more detailed representations of human tissues, including bones, muscles, and internal organs, which affect EMF interaction and the resulting E values. In contrast, the phantom used in the experiments has a less complex structure, leading to smaller differences in the E values.

The bones included in the simulation models can be a significant factor influencing the distribution of EMF in the human body. Part of the radiation can be reflected from the bones, which leads to the formation of a local maximum on the body surface.

The near field of EM radiation at very short distances (up to a few millimetres) is complex and unpredictable. The relationships used to convert power levels (E values) are primarily intended for the far field, which could cause inaccuracies at distances as small as 5 mm. The size of the deviations decreased at larger distances (15 mm, 50 mm, and 300 mm), where they were significantly smaller between measurements and simulations.

To reduce the computational and time requirements, the simulations used a simplified, hollow model of the pacemaker. However, the real experiments used a fully functional pacemaker with all the complex components such as batteries, electronic circuits, and sensors. We think that the difference in complexity and material composition could have also affected the resulting values

The following graphical representation (Figure 12) provides a detailed analysis and comparison of the results obtained from simulations and experimental measurements for situations where a PM was implanted and for situations without a pacemaker. The graph clearly shows the differences in E between individual models and scenarios.

When comparing individual scenarios of simulations and measurements, we found that the average increase in E values in simulations with an implanted device for different distances was as follows: at a distance of 5 mm by 23.47%, at a distance of 15 mm by 37.14%, and at a distance of 50 mm by 62.36%. It is obvious that at larger distances (e.g., 50 mm) the EMF is less concentrated and more scattered. A pacemaker, as a complex object with different materials and components, could thus significantly affect the distribution of the field. This effect is likely to be more pronounced at longer distances, where various interactions, reflections, scatterings, or absorptions can occur.

One of the factors that influences the results is tissue heterogeneity, since EMF can propagate through different tissues with different electrical properties. These tissues combined with the implant would thus create a complex system where each element could contribute to the overall field intensity. A distance of 50 mm could represent a sufficient space where these interactions would be more pronounced.

Expert sources such as the European Society of Cardiology [31] stress that it is safe to maintain this distance when using mobile phones and also recommend using the phone on the ear opposite to where the pacemaker is implanted.

Based on our simulations and experimental measurements, we found that maintaining a minimum distance of 15 cm between the mobile phone and the pacemaker is not sufficient to ensure patient safety. Our results show that even at a distance of 30 cm, the E values for the models were: MM—54.96 V/m, for Child—65.34 V/m, for Laura—62.81 V/m, and measurements—52.72 In M. In the case of the Child and Laura complex models, the E values still exceeded the exposure limit of 58 V/m for a frequency of 1800 MHz. Our recommendation is to push this limit up to more than 30 cm.

Another recommendation is protective equipment that reduces the exposure of parts of the human body to EMF. However, these aids must have a certificate of satisfactory EM compatibility.