Novel technology at hand to measure skin hydration by Biodisplay smartphone touch screen panel

This study was a prospective clinical trial that was approved by the institutional review board of Samsung Medical Center, Sungkyunkwan University School of Medicine in Seoul, Korea (2020-04-028). This study has been registered in Clinical Research Information Service on June 18, 2020 (http://cris.nih.go.kr, KCT0005146). The study was conducted according to the protocols of the Declaration of Helsinki.

Study populations

The participants in this study were enrolled based on the following inclusion and exclusion criteria: Subjects who were between the ages of 20 and 60 years, and who were familiar with using smartphone applications. Subjects were excluded if they had a diagnosis of eczema or hyperhidrosis, or a history of any disease requiring treatment at the evaluated sites. All the included patients signed written informed consent for their participation.

Devices

Biodisplay

A smartphone (Galaxy S9, SM-G960N, Samsung Electronics, Suwon, Korea) and Biodisplay (Samsung Display, Yongin, Korea) were used to examine skin hydration (Fig. 2). Changes in capacitance occur in the skin because the human body is mainly composed of water, and the epidermis of the skin is optimized for moisture storage with its specific structure, similar to bricks and mortar¹⁶. The dielectric constant of water is much higher than that of air, and human skin can highly influence the capacitance variance^17,18. Based on this principle, the Biodisplay can measure capacitance based on skin moisture. When smartphone users touch their skin onto the touch screen panel, the capacitance is measured at the contact area. The output data were analog to digital converter code values on each pixel, so in this experiment, 528 values in total were presented by a matrix. The Biodisplay algorithm consists of valid capacitance sampling, data mining, and the conversion of capacitance to skin moisture. Valid capacitance data were extracted by applying a threshold in this algorithm, then used to calculate the portion of the touched area. Finally, the capacitance value was applied to a specific equation that converted the capacitance into a measure of skin moisture (Fig. 3).

Scheme for skin hydration measurement using smartphone.

Data processing flow for skin hydration measurement using smartphone.

** cap.ave: Average of cap code at the contact site.

Biodisplay has been developed with a measurement range of 1 to 100 a.u. and the coefficient of variation (C.V.) of less than ± 10%. The evaluation was conducted to satisfy the specifications when the pressure and temperature change. Biodisplay varied within 3 a.u according to the pressure for all subjects. There was a tendency to vary within the range, which satisfied the C.V. ± 10% specification. When it comes to the influence of temperature on the measurement of the Biodisplay, the maximum of difference was 2 a.u. depending on the temperature. Therefore, the effect on the temperature is very small at a level that satisfied the C.V. ± 10% specification (Supplementary file 1).

Hydration probe (HP)

For comparison as a preexisting objective skin hydration measuring device, the HP of DermaLab Combo was used as a gold standard method to measure skin hydration (μS). This HP uses the conductance principle to measure skin hydration. It has a central circular electrode surrounded by eight small pins.

Procedure

The study was conducted between May 2020 and June 2020. All measurements were performed in a room in which the temperature and the humidity were controlled ((20–22) °C, (45–55) %). Participants were asked to rest for 30 min before the procedure. The skin hydration was measured by two skillful specialists using HP to minimize the influence of investigators like pressure during measurements and the Biodisplay was applied by each participant using their dominant hand. Generally, smartphone users grab their smartphone with their dominant hand and therefore, we measured the skin hydration on the defined point on the arm, nondominant volar arm, 5 cm above the wrist, considering the real setting of measurement using smartphone. All measurements were repeated three times by each device.

First, participants took hold of the smartphone, and made contact between their forearm and the smartphone panel. As the participants maintained contact between their forearm and the smartphone panel, capacitance data on the contact area were recorded for 2 to 3 s, which was the required sensing time. When the measurement was complete, an alarm went off. The skin hydration values were calculated and presented on the mobile display (Fig. 2). All measurements were repeated three times. Then, we measured skin hydration on the forearm three times using HP.

Next, participants applied two fingertip units of moisturizer onto their non-dominant volar forearm, and waited 30 min to allow for absorption. Prior to all measurements, any residual moisturizer should be gently removed through light wiping with a non-woven tissue to ensure the exact evaluation of skin hydration¹⁴. The Biodisplay is expected to be useful in everyday life, which means people may use it on their skin without washing the measurement area, and they may even use it on their skin over makeup. Therefore, we measured skin hydration both before and after wiping off all residual moisturizer. After 30 min of application, the skin hydration was measured by the Biodisplay and HP three times each in the same manner. Finally, any remaining moisturizer was wiped off from the skin, and the skin hydration was again measured by the Biodisplay and HP three times each.

Statistical analysis

Statistical analysis was executed using R 4.1.0 (Vienna, Austria; http://www.r-project.org/). The test–retest reliability was assessed using intraclass correlation coefficients (ICC). ICC values were interpreted as (0–0.50), poor agreement; > (0.50–0.75), moderate agreement; > (0.75–0.90), good agreement; and > (0.75–0.90), excellent agreement¹⁹. A high ICC means that the measurement can be performed reliably by the same person multiple times.

To evaluate the correlation between Biodisplay and HP, Shapiro–Wilk normality test was done first, and Pearson’s correlation analysis (PCC) was performed along with scatter plot²⁰. After, PCC was calculated, and p-values < 0.05 were considered significant. The range of values for PCC is between (−1 and + 1) (total negative correlation and total positive correlation), where zero values mean no correlation at all. PCC was interpreted as < 0.30: negligible correlation, low positive(negative) correlation (0.30 to 0.50), moderate positive(negative) correlation (0.50 to 0.70), high positive(negative) correlation (0.70 to 0.90) and ≥ 0.90: very high positive (negative) correlation²¹. All statistical analyses were conducted by two biostatistics specialists (SW Kim and JS Shim).