Electromagnetic Sensing Techniques for Monitoring Atopic Dermatitis-Current Practices and Possible Advancements: A Review

Abstract

Atopic dermatitis (AD) is one of the most common skin disorders, affecting nearly one-fifth of children and adolescents worldwide, and currently, the only method of monitoring the condition is through an in-person visual examination by a clinician. This method of assessment poses an inherent risk of subjectivity and can be restrictive to patients who do not have access to or cannot visit hospitals. Advances in digital sensing technologies can serve as a foundation for the development of a new generation of e-health devices that provide accurate and empirical evaluation of the condition to patients worldwide. The goal of this review is to study the past, present, and future of AD monitoring. First, current medical practices such as biopsy, tape stripping and blood serum are discussed with their merits and demerits. Then, alternative digital methods of medical evaluation are highlighted with the focus on non-invasive monitoring using biomarkers of AD-TEWL, skin permittivity, elasticity, and pruritus. Finally, possible future technologies are showcased such as radio frequency reflectometry and optical spectroscopy along with a short discussion to provoke research into improving the current techniques and employing the new ones to develop an AD monitoring device, which could eventually facilitate medical diagnosis.

Keywords: atopic dermatitis; electromagnetic sensing; flexible wearable sensors; interdigitated capacitive sensor; near-infrared range spectroscopy; neural networks; non-invasive monitoring; radio frequency reflectometry; telemedical sensors; transepidermal water loss.

Figure 1

Methods of medical examination of AD and their corresponding potential biomarkers. The three main empirical methods are listed, all of which are performed under a controlled clinical environment. Figure created with BioRender.

Figure 2

Tape-stripping procedure illustrated with pictures. (A) presents the top and bottom view of the strip and (B) demonstrates the activation process used to extract the biomarkers. Image reproduced with permissions from [33].

Figure 3

AquaFlux evaporimeter and its measurement principle. The chamber visible on the right is housed within the tip of the AquaFlux evaporimeter and contains the humidity sensors that detect the TEWL. Image reproduced with permissions from [47].

Figure 4

Images of multimodal device for measurement of TEWL, skin conductance and skin hardness constructed by Sim et al. A photograph is visible in (a) (left) and in (b) (right), its operating principle is presented. Images reproduced with permission from [50].

Figure 5

Measurement principle of the Corneometer^® CM 825. On the top is the front view of the electrode, revealing its interdigitated shape and on the bottom is the side view, illustrating the scattered electric field. Image reproduced with permission from [64].

Figure 6

Images of the textile-based impedance sensor created by Jang et al. (a,b) present a schematic on the top (a) and side (b) view of the sensor, respectively, and (c) demonstrates its flexibility through photographs. Images reproduced with permission from [70].

Figure 7

Examples of actigraphy sensors for scratch detection. (a–c)—Photograph, schematic and flow chart of the acoustic scratch detection sensor by Noro et al.; (d) photograph of the MicroMini Motionlogger^® to showcase its compactness and non-invasiveness. Images reproduced with permissions from [22,78].

Figure 8

Schematic of the EczemaNet model created by Pan et al. The model can identify photographs of AD and predict the severity of the disease. Image reproduced with permissions from [83].

Figure 9

Illustration of different non-invasive telemedical methods for monitoring AD in vivo. The respective measurand parameters are presented on the top of each method’s illustration. The TEWL method features an enclosed chamber with a relative humidity sensor inside so that water vapors from the skin affect its reading. The skin permittivity sensor consists of 2 or more electrodes in contact with the skin of which one is a signal source and the other one is a signal pathway (ground) to establish a scattered electric field through the skin. The skin elasticity sensor uses negative pressure to suction parts of the skin and emit light through it to test the skin’s properties. The scratching frequency sensor detects when a person is scratching their AD lesions. The neural network imaging system uses a camera to distinguish lesions of AD from other similar conditions such as melanoma, skin burn, and psoriasis.

Figure 10

Experimental setup and results of RF reflectometry measurements performed by Schiavoni et al. (a) Presents a photograph of the setup with the VNA used to perform the TDR calculations and (b) measurement of s-parameters for four cases of forearm hydration. Images reproduced with permission from [90].

Figure 11

Schematic of handheld in vivo CRM system utilized by Dev et al. The system is housed within the handheld probe visible in the top right corner—the cable on the left is connected to the laser source and the cable connecting to the spectrograph is not illustrated. Image reproduced with permissions from [96].

Figure 12

Absorbance spectrum of water and porcine skin in the NIR range. The absorbance peaks around 1400 nm and 1900 nm match almost perfectly, indicating that the absorbance response in the skin is due to the water concentration. Image reproduced with permissions from [13].

Figure 13

Schematic of opto-electronic patch sensor created by Yan et al. Four channels of NIR LEDs are positioned around a single PD cell. The channel can emit light with varying wavelengths to achieve a higher sensitivity. Image reproduced with permissions from [105].