**3. Electromagnetic compatibility for implantable medical device**

### **3.1 EMI shield**

The EMI shield is designed to decrease the electromagnetic (EM) wave transmission using a shield to increase the reflection or absorption of EM wave incident at the interfaces between different mediums. As shown in **Figure 5**, the electric and magnetic fields of EM waves are perpendicular to each component and the EM

*Electromagnetic Compatibility Issues in Medical Devices DOI: http://dx.doi.org/10.5772/intechopen.99694*

**Figure 5.**

*Time-varying EM waves and the propagation, transmission, and reflection of the EM waves in different mediums.*

propagation direction, which can be expressed as phasor form [23], according to Eqs. (1) and (2). Where γ, α, β are the propagation, attenuation, phase constants of the medium, respectively; E0 and H0 are the amplitude of the electric and magnetic fields.

$$\mathbf{E} = \hat{\mathbf{a}}\_{\text{Y}} \mathbf{E}\_{0} \mathbf{e}^{-\text{yr}} = \hat{\mathbf{a}}\_{\text{Y}} \mathbf{E}\_{0} \mathbf{e}^{-\text{oz}} \mathbf{e}^{-\text{j0z}} \tag{1}$$

$$\mathbf{H} = \hat{\mathbf{a}}\_{\text{\textquotedblleft}} \mathbf{H}\_{0} \mathbf{e}^{-\text{\textquotedblleft}\mathbf{z}} = \hat{\mathbf{a}}\_{\text{\textquotedblleft}\mathbf{H}} \mathbf{e}^{-\text{\textquotedblleft}\mathbf{z}} \mathbf{e}^{-\text{\textquotedblleft}\mathbf{z}} \tag{2}$$

The EM wave propagates at the interface of two different mediums that induce reflection due to impedance mismatching. The reflection coefficient (R12) and transmission coefficient (T12) at the interface between two mediums can be determined, according to Eqs. (3) and (4). Ei, Er, η1, η<sup>2</sup> represent incident, reflected electric fields, impedances in Medium 1 and Medium 2, respectively. The impedance in the medium can be defined by a ratio of electric field and magnetic field, which is related to the permittivity (ε), permeability (μ), and conductivity (σ), according to Eq. (5).

$$\mathbf{R\_{12}} = \frac{\mathbf{E\_r}}{\mathbf{E\_i}} = \frac{\eta\_2 - \eta\_1}{\eta\_2 + \eta\_1} \tag{3}$$

$$\mathbf{T\_{12}} = \frac{\mathbf{E\_t}}{\mathbf{E\_i}} = \frac{2\eta\_2}{\eta\_2 + \eta\_1} \tag{4}$$

$$\eta = \frac{|\mathbf{E}|}{|\mathbf{H}|} = \sqrt{\frac{\text{joļņ}}{\sigma + \text{joē}}} \tag{5}$$

The time-domain electric field can be rewritten as Eq. (6), according to Eq. (1).

$$\mathbf{E} = \hat{\mathbf{a}}\_{\mathbf{y}} \mathbf{E}\_{0} \mathbf{e}^{-\text{yr}} = \hat{\mathbf{a}}\_{\mathbf{y}} \mathbf{E}\_{0} \mathbf{e}^{-\text{oz}} \cos \left( \alpha \mathbf{t} - \beta \mathbf{z} \right) \tag{6}$$

When the EM wave propagates in the conductive shield (loss medium) at a time of zero, the expression between the distance and amplitude of the electric field can be obtained, according to (7).

$$\mathbf{E} = \hat{\mathbf{a}}\_{\text{y}} \mathbf{E}\_{0} \mathbf{e}^{-\alpha \mathbf{z}} \cos \left( \beta \mathbf{z} \right) \tag{7}$$

**Figure 6** demonstrates the EM wave propagation in the conductive shield. The skin depth (δ) can be defined as that penetration distance at which the intensity of

**Figure 6.** *Schematic of skin depth of EM waves using the shield.*

the electric field attenuates to 1/e of the original incident wave intensity. According to the Eqs. (8) and (9), the skin depth (δ) is the reciprocal of attenuation constant (α), which is related to the EM operating frequency, permeability (μ), and conductivity (σ) in the medium [10].

$$\frac{\mathbf{E}\_0}{\mathbf{e}} = \mathbf{E}\_0 \mathbf{e}^{-\alpha \mathbf{8}} \tag{8}$$

$$\delta = \frac{1}{\alpha} = \frac{1}{\sqrt{\pi f \mu \sigma}} \tag{9}$$

Thus, conductivity and permeability in shielding design play an important role in EM wave absorption enhancement, thus increasing the overall EMI shielding effectiveness.

However, the implantable medical device is not a fully closed system that must require the openings of the shield to interact with external equipment such as body sensing devices for signal transmit or receive. In some cases, the external controlled magnetic fields or electrical signals can be utilized to externally modulate the stimulation protocol of implantable medical devices according to patients' clinical requirements. So, the selective filtering of EMI waves is important for implantable medical devices to classify the noise and external signals. Thus, the EMI filter was provided in the following subsection to promote the EMC applications in implantable medical devices.

#### **3.2 EMI filter**

The filter implementation is also a strategy for EMI elimination in medical devices [24–27]. For example, the typical ranges of P, R, T waves in ECG are 20 to 40 Hz, 18 to 50 Hz, and 0 to 10 Hz [28], as shown in **Table 2**.

Filtering can be divided into active and passive modes. Active filters consist of several operational amplifiers and passive elements such as capacitors and resistors. [29–32]. The active filters are applied in wide applications owing to excellent filter performance. However, the active filters need a power source to sustain the

*Electromagnetic Compatibility Issues in Medical Devices DOI: http://dx.doi.org/10.5772/intechopen.99694*


**Table 2.**

*Significant frequency bandwidth of ECG waveform [28].*

**Figure 7.** *Filter implementation for removing the time-varying electric and magnetic fields.*

operations. Moreover, the upper frequency of active filters may be limited. Thus, the active filter is not suitable for EMI filtering in implantable medical devices. The active filter can perform programmatical filtering for the received signals, thus separating the signal from noise. However, programmatical computation requires a high-cost and complex circuit with larger power consumption to sustain the processing functions. Because the implantable devices aim to sustain life, such devices are not expected to remove or insert frequently because of extremely high costs for device failure.

Moreover, it is not easy to replace it if the devices fail due to the high risk of surgery. The concern regarding the surgery risk, which makes battery life issues more important. Owing to the battery requirements of implantable medical devices, the minimization of filters' power is crucial to prolong the implantable device lifespan.

The capacitor-based passive filter is frequently used for most high-frequency noise in the surrounding ambient for the filter design regarding implantable medical devices. Capacitors can filter EMI noise utilizing absorption and smoothing of electromagnetic noise. The high-frequency noise attenuates as quickly as charging and discharging the capacitor-based filter. Absorbing such EMI noise to the ground will neutralize or prevent specific frequencies from passing through the circuit, as shown in **Figure 7**.

The discoidal capacitor and feedthrough capacitor array were commonly utilized in the practical applications of medical devices, which deliver high-density performance with low-volume packaging [33–35], as shown in **Figure 8**.

The circular-shaped discoidal capacitor is one of the most common constructions for feed-through-style EMI filters. Circular capacitors outside and inside diameters serve as connection points for the case and the lead and serve as the capacitor poles. Moreover, several discoidal capacitors can be assembled to integrate as a capacitor array on a single piece of ceramic [36]. Such assembly offers the highest filter performance within the limited physical dimension. Thus, the feedthrough

**Figure 8.** *The feedthrough capacitor array and discoidal capacitor for the implantable medical device.*

capacitor array provides the merits of the miniature dimension and lightweight within a high-density implantable device [37].

However, a feedthrough filter will have a double impact on battery life. First, a minimal amount of current always flows between the plates of the charging capacitor. Since one capacitor is processing the signal and the other capacitor is grounded, the leakage current will drain the battery over time. A strong dielectric with an appropriate thickness can resist this current flow, thereby significantly reducing battery consumption. Besides, filter design in implantable medical devices is to minimize the loss of expected signals. The filter's insertion loss implies how much a signal will be lost or reduced for each frequency. An excellent filter requires a lower insertion loss for signal frequencies and a higher insertion loss for noise frequencies. Some energy in the expected signal will attenuate in internal resistance and inductance of the filter, which implies that the implantable battery needs optimize in the power design. Thus, an optimized design can suppress the energy loss. The less power dissipated in the battery, which extends battery life and improves effectiveness.
