**5. Testing method of dry electrodes**

This section reviews the existing methods used for developing and testing the newly designed dry electrodes. The existing methods has been broadly classified into *in vivo* and *in vitro* type of testing. *In vitro* testing of dry electrodes refers to the evaluation of the performance of wearable electrodes across the human subjects, and *in vitro* testing refers to evaluating the electrode performance on a platform (aritifical skin) that has similar properties to that of human skin.

#### **5.1** *In vivo* **testing**

Historically, the development and testing of electrodes are empirically performed by conducting experiments *in vivo* on human volunteers. This method of testing has two limitations: First, the properties of the human skin change over time, which leads to changes in the skin impedance, thus making it challenging to perform reproducible measurements [24]. Secondly, the properties of human skin differ for each person. Hence, the intra and inter-subject variability complicate the understanding of the performance of the dry electrodes in the acquisition of biopotential signals. Therefore, the performance of electrodes cannot be investigated and the role of various electrode parameters including electrode material, and applicability of the electrode design to a wider range of populations, cannot be clearly understood.

#### **5.2** *In vitro* **testing**

To overcome the limitations of *in vivo* testing, some efforts have been made to fabricate a synthetic model of skin and tissue. These so-called *phantoms* simulate the electrical properties of tissues. The phantoms can be used as controlled benchtop

### *Factors Affecting Wearable Electrode Performance and Development of Biomimetic Skin… DOI: http://dx.doi.org/10.5772/intechopen.111429*

testing platforms and can be used to facilitate the development of the electrode designs as there is no explicit control on human variables.

**Figure 5** shows the electrical properties of two layers of the skin, which comprise the topmost layer known as the stratum corneum, and all the inner layers of the skin other than the stratum corneum, which are considered dermis/deeper tissue layers. The electrical properties of the stratum corneum represented in **Figure 5** were investigated in one of the initial research works by Yamamoto et al. [44], where the stratum corneum was removed from a forearm of a human subject with a cellulose adhesive tape. The keratin layers of the stratum corneum were removed stepwise by stripping the skin with the cellulose tape and the electrical properties of the dermis and stratum corneum were studied. The impedance of the stratum corneum accounts for a major portion of the whole impedance measured at the skin, and it dominates in the lowfrequency range between 1 Hz and 10 kHz. The region of interest in electrophysiological signals is low-frequency regions, and the stratum corneum impedance plays the dominant role in the low-frequency region. In this section, the existing phantoms have been broadly classified into single-layered and two-layered phantoms.

#### *5.2.1 Single-layered phantom*

Skin phantoms, also known as artificial/synthetic skin, are made using different materials to mimic various kinds of skin properties. Skin properties can be broadly categorized into surface, mechanical, acoustic, optical, electrical, and thermal properties. For mimicking electrical properties, gelatinous substances and elastomers are mostly used [45]. Several researchers have fabricated skin phantoms using gelatinous substances such as gelatin [24] and agar [15, 46]. The agar-based phantom's electric and dielectric properties can be adjusted by adding NaCl and polyethylene powder, respectively. Using this approach, Ito et al. fabricated the conventional electromagnetic phantom as a whole layer and used it to mimic electrical properties for higher frequency ranges (300 MHz–2.5 GHz) [47]. Later on, Yamamoto et al. used the above approach and further tuned electrical properties by adding the carbon microcoil to

#### **Figure 5.**

*Electrical properties of two layers of skin, where ρ<sup>k</sup> and ε<sup>k</sup> are average resistivity and dielectric constants of stratum corneum; ρ<sup>c</sup> and ε<sup>c</sup> are average resistivity and dielectric constants of inner layers (dermis) (Image reproduced from reference [44]).*

increase the relative permittivity of the phantom and thus fabricated a phantom that simulated electrical properties ranging between 1 MHz up to 2.5 GHz. However, the hydrous phantoms could not be further tuned to mimic the electrical properties for frequencies below 1 MHz [48]. Thus, the conventional approach using agar, DI water, NaCl, and polyethylene simulate the electrical characteristics of a high-content water tissue such as muscle and brain, and is incapable of simulating the electrical properties of the skin.

In another such work, Kalra et al. fabricated a single-layer phantom using oil in gelatin for simulating dielectric properties in a low-frequency range (20 Hz– 300 kHz). However, the results are four orders of magnitude away from the desired skin dielectric properties in the low-frequency range [49]. A similar approach using gelatin was followed for fabricating a phantom for electrophysiology in the recent work of Owda et al. A single-layer phantom was made, and efforts were made to tune the electrical properties with different gelatin and NaCl concentrations [50]. The contact impedance profile was compared for the developed gelatin-based phantom and ex vivo porcine skin over 20–1000 Hz [50]. However, the impedance of human skin is much higher than that of porcine skin. For biopotential signals, the stratum corneum impedance is of interest as it dominates the skin impedance in the lowfrequency range (1 Hz to 10 kHz) [51]. Therefore, a single-layer phantom approach does not include the effect of the outer layer of skin. Hence, there is a need for a twolayered skin phantom.
