1.

Introduction

Optical tissue phantoms (OTPs) that simulate optically compatible tissue characteristics have been used for the evaluation of optical system performance,¹ calibration of optical devices,^1–3 and in vitro simulation of light distribution.^2,3 Unlike other types of OTPs (gel or liquid), solid OTPs have been preferentially used because they have a long shelf life and can be easily produced.^3–5 However, it is difficult to build thin-layer solid OTPs (TSOTPs) that are of comparable thickness to the human epidermis with several layers (four to five layers).⁶

Various methods have been developed to build OTPs with epidermal thickness.^7–9 Bruin et al.¹⁰ and Saager et al.⁴ described the use of a customized frame to build OTPs with epidermal thickness, and Urso et al.⁷ reported that epidermal thickness was achieved by placing an OTP mixture between glass slides. Furthermore, Tseng et al.¹¹ reported that epidermal thickness could be adjusted by calculating the solvent volume before the curing process, while Bergmann et al. reported that the thickness of an OTP could be controlled by adjusting the repetition rate of the spraying process.⁶

Although many materials and methods have been proposed for the construction of TSOTPs with epidermal thickness, there remains substantial inaccuracy and inconvenience among these procedures, and there has been no consensus on an appropriate and efficient procedure for the fabrication of TSOTPs.⁶ The present methods have several common limitations such as long production times, precipitation of chromophore particles, and microscopic particle clustering.^4,7,10 Moreover, none of these studies have successfully achieved the epidermal thicknesses of $\le 10{\mu \text{m}}^2$.¹²

This study introduces a new spin-coating method (SCM) to fabricate reliable and reproducible TSOTPs with epidermal thickness. SCMs have been widely used in the microelectronics industry to fabricate thin and homogeneous films over large areas.^13–19 In addition, it provides simple and quick process without any cumbersome processes. In this study, the SCM controls the TSOTP thickness as a function of the spin speed and number of layers. TSOTPs were fabricated to mimic the epidermis layer only considering the absorption property for which various concentrations of Indian ink were used. Finally, the TSOTPs were quantitatively evaluated to determine the absorption coefficients and the thickness.

2.

Materials and Methods

Figure 1(a) shows the procedure used for the preparation of the OTP mixture. Figure 1(b) shows the procedure used to characterize the TSOTPs, which included the preparation of the OTP mixture, the spin coating process, thickness measurements of the TSOTPs, and measurements of their optical properties.

Fig. 1

(a) Procedure for preparation of the OTP mixture; (b) procedure for characterization of the TSOTP, which is composed of the OTP preparation, spin-coating process, thickness measurement of OTP, and optical property measurement.

3.

Results

The top, center, and bottom frames in Fig. 2 show digital microscopic images of one-layer, two-layer, and three-layer TSOTPs, respectively, fabricated at 250 rpm [Fig. 2(a)], and 2500 rpm [Fig. 2(b)]. The layers of TSOTP were of homogeneous thickness in the region of interest (ROI), and their thickness increased as a function of the number of layers. The boundaries between the TSOTPs and the glass slides were clearly distinguished. The scratch on the cross-sectional TSOTP, as observed in the figure, might be caused in the cutting process of TSOTP when a rolling diamond cutter is used for the thickness measurement; this scratch does not affect the phantom thickness and homogeneity.

Fig. 2

Sample images of TSOTPs fabricated at spin speeds of (a) 250 rpm; (b) 2500 rpm; the top, center, and bottom images indicate one-layer, two-layer, and three-layer TSOTPs, respectively.

Figure 3 shows thickness variation as a function of the spin speed and absorber concentration. The TSOTP thickness decreased exponentially as a function of the spin speed and drastically decreased up to 750 rpm. One-layer TSOTPs reached a maximum thickness of $65\pm 0.28\mu \text{m}$ at 250 rpm and a minimum thickness of $5.1\pm 0.17\mu \text{m}$ at 2500 rpm. The five different concentrations of absorber did not affect the thicknesses of TSOTPs fabricated at the same spin speeds.

Fig. 3

Thickness variation of one-layer TSOTPs as a function of spin speeds from 250 to 2500 rpm at five different absorber concentrations.

Figure 4 shows the thickness variations of multilayer TSOTPs as a function of the spin speed at the same absorber concentration (1.0% of total volume). The TSOTP thickness decreased exponentially as a function of the spin speed and increased as a function of the number of layers. Mathematical equations (Table 1) were obtained to predict TSOTP thickness as a function of the spin speed by performing a power function curve fitting (solid line in the figure) as follows:

Fig. 4

Thickness variation of multilayer TSOTPs as a function of spin speeds. Power curve fitting shows correlation coefficients of $R^2=0.95$ for one layer, $R^2=0.99$ for two layers, $R^2=0.99$ for three layers.

Table 1

Power curve fitting equations for thickness variation of TSOTPs as a function of spin speed.

Figure 5 shows that the absorption coefficient exponentially increased as a function of the absorber concentration. The following mathematical equation was extracted by performing a power function curve fitting:

Fig. 5

Absorption coefficients (${\mathrm{mm}}^{-1}$) of TSOTPs measured at 660 nm as a function of absorber concentrations from 0.2% to 1.0%.

Figure 6(a) shows digital microscopic images of the dermal OTP (left image) and the double layer OTP (right image) that combines epidermal TSOTP and dermal OTP. The boundary between the epidermal TSOTP and dermal OTP was clearly distinguished in the double layer OTP (right image). Figure 6(b) shows the transport attenuation coefficients (${\mu }_{\mathrm{tr}}$ of three double-layer OTPs. Each double layer OTP presented the average ${\mu }_{\mathrm{tr}}$ of 4.78, 3.91, and $4.44{\mathrm{mm}}^{-1}$ and the uniform ${\mu }_{\mathrm{tr}}$ in the ROI, as indicated by low standard deviations.

Fig. 6

(a) Sample images of the dermal optical tissue phantom (OTP) (left image) and the double-layer OTP (right image), which has epidermal thin-layer solid OTP and dermal OTP; (b) transport attenuation coefficients (${\mathrm{mm}}^{-1}$) of double layer OTPs measured at 660 nm. The ${\mu }_{\mathrm{tr}}$ of the double-layer OTPs was 4.78, 3.91, and $4.44{\mathrm{mm}}^{-1}$, respectively, thus having an average of $4.38{\mathrm{mm}}^{-1}$.

4.

Discussion

Although conventional solid OTPs have been routinely used as essential tools in biomedical optics, the control of their thickness is difficult, requiring cumbersome and time-consuming production procedures. This study introduced a new SCM to partially address these problems, thus improve the fabrication efficacy of OTPs. When compared with conventional methods, the SCM can reduce fabrication time, easily control OTP thickness by managing the spin speed, and fabricate multilayer OTPs. The SCM can be used to easily fabricate OTPs of various thicknesses by managing the number of layers and the spin speed. In the conventional method, it is difficult and cumbersome to replicate the thickness of the stratum corneum ($\le 10\mu \text{m}$) in the epidermal layer.²² However, the SCM was successfully used to fabricate TSOTPs with a minimum thickness of 5.1 μm and a maximum thickness of approximately 65 μm for a one-layer coating, which is comparable with the epidermal thickness (40 to 150 μm). The epidermis is composed of four to five layers depending on the skin region. By using the SCM, TSOTPs with epidermal thickness can be fabricated with one-layer or multilayer coating depending on the application. In future studies, the additional parameters of the SCM that may also affect thickness should be optimized with respect to the fabrication materials and procedures of OTP.

The relationship between the spin speed and the thickness demonstrated that the thickness of the TSOTP may be affected by the centrifugal force associated with the rotary motion. A faster rotary motion generates greater centrifugal force, resulting in decreased thickness.^13–17 Such a mechanism might affect the thickness of the TSOTP, which exponentially decreases as a function of the spin speed (Figs. 3 and 4).

In the case of one-layer TSOTP, absorption coefficients increased as a function of absorber concentration (Fig. 5) but did not affect the thickness of the TSOTP (Fig. 3). It is known that the viscosity of base material is strongly related to the solvent of evaporation.¹³ This phenomenon results in a difference in thickness. In our case, there is no solvent evaporation process, owing to the properties of nonvolatile epoxy. Accordingly, the different absorber concentration may not directly affect the thickness of the TSOTP, as shown in Fig. 3.

The thickness of the TSOTP might be affected by the viscosity of the OTP mixture. In a previous study,¹³ the film thickness was affected by the viscosity of the base solution. Based on that study, the thickness of the TSOTP was investigated for the viscosity of the OTP mixture. In the experiment, epoxy had a high viscosity and was not uniformly distributed on the glass slides during the spin coating process; therefore, a viscosity depressant was used to reduce the viscosity of the TSOTP. The viscosity of the OTP mixture was decreased significantly and therefore directly affected the uniformity of the OTP thickness. In future studies, it may be beneficial to investigate the effects of viscosity of the OTP mixture on the TSOTP thickness.

As in the case of the one-layer TSOTPs, the thickness of the multilayer TSOTPs was exponentially decreased as a function of the spin speed and increased as a function of the number of layers at all spin speeds (Fig. 4). These results show that the SCM can fabricate TSOTPs of various thicknesses by manipulating the spin speed and the number of layers. In multilayer TSOTPs fabricated at the same concentration, it was difficult to identify the boundary between each TSOTP layer. This may arise from the use of the same absorption property, no air gap between layers, or the very low thickness of each layer. This study presented only three layers of TSOTP with the same absorption property to prove the feasibility of the SCM in the fabrication of the multilayer TSOTP. In future study, an epidermis layer with four to five layers will be fabricated with a different absorption property and thickness.

Double-layer OTPs with the same absorption and scattering properties presented stable ${\mu }_{\mathrm{tr}}$ as indicated by the low standard deviations [Fig. 6(b)]. In performing IAD, tissue optical properties were identified using various approaches in the previous studies because the optical properties of human tissues vary depending on the anatomical locations, skin color, and optical agents; moreover, these properties are not yet established for all skin layers.^20,21

This study first investigated the feasibility of the SCM in the fabrication of the epidermal TSOTP considering only absorption property and then double-layer OTP, which has epidermal TSOTP, and dermal OTP was fabricated by considering absorption and scattering properties. It should be noted that the results may vary depending on the fabrication material and procedure of the OTP.

5.

Conclusions

The SCM can be used to quantitatively fabricate one-layer and multilayer TSOTPs by controlling the spin speed to achieve epidermal thickness with absorption property. In addition, a double-layer OTP was fabricated by combining the dermal OTP onto the epidermal TSOTP fabricated using the SCM. In future studies, the epidermal TSOTP will be fabricated with skin texture (wrinkles) and with multiple layers having different absorption properties. The method of fabricating double-layer OTP will be characterized in detail.