Hyperspectral assessment of acne skin exposed to intense pulsed light (IPL) intense pulsed light in acne treatment

Abstract

Background

The mechanism of intense pulsed light action on the skin is based on selective photothermolysis. The light delivered to the tissue is scattered and absorbed by chromophores that absorb a beam of radiation of a specific length. The skin reflectance changes depending on the physiological state of the tissue, as shown by the hyperspectral camera. The aim of the study was to assess the hyperspectral reflectance of acne skin before and after intense pulsed light (IPL) therapy and to compare it with the reflectance of healthy skin.

Materials and methods

The study involved 27 volunteers with diagnosed moderate acne. The control group consisted of 20 people without acne lesions. All acne volunteers underwent a series of four treatments using IPL at weekly intervals. The volunteers with acne lesions were photographed before the series of treatments and a week after the 4th treatment.

Results

Acne skin shows lower reflectance than healthy skin. Acne skin after IPL therapy is characterized by a higher reflectance compared to acne skin before the therapy and resembles the reflectance of the skin of the control group. A statistically significant difference was found between the acne skin before the treatments and the skin of the control group.

Conclusions

The effect of IPL therapy on acne skin is the increase of its reflectance by reducing the number of chromophores, which brings it closer to the reflectance value of healthy skin. Hyperspectral imaging allows for: the evaluation of the treated skin at each stage, a precise selection of the light wavelength depending on the problem, and therefore, for optimizing the number of irradiations and increasing the safety of the therapy.

1. INTRODUCTION

The intense pulsed light (IPL) device is a tool commonly used in dermatology. ¹ Unlike laser systems, IPL lamps emit high‐intensity polychromatic, incoherent and diffused light covering a wide spectrum of wavelengths between 420 and 1200 nm. These properties allow for a precise selection of individual treatment parameters and adaptation to different skin types and various skin problems. It is used to reduce discoloration, ² erythema, acne, ³ or photorejuvenation. ⁴

The mechanism of action of IPL is based on selective photothermolysis. ⁵ The light penetrates the skin until it reaches the desired chromophore. Due to the different maximum absorption values of the respective target structures, appropriate wavelengths can be selected to purposefully heat and coagulate them. Endogenous or exogenous chromophores in the skin absorb photons, releasing thermal energy that heats them, leading to the destruction of the target tissue by thermocoagulation. ⁶ , ⁷ , ⁸

The primary chromophores in the skin that make IPL acne therapy possible are hemoglobin, melanin, water, and porphyrins. Hemoglobin absorbs light mainly at a wavelength of 580 nm, and melanin absorbs the entire visible spectral range from 400 to 750 nm. Porphyrins—products of the metabolism of Cutibacterium acnes—absorb some blue light (400–450 nm), with the most efficient absorption being the wavelength of 415 nm, and to a lesser extent in long visible bands such as orange (590–635 nm) and red light (635–700 nm). ⁹

The tissue imaging with a hyperspectral camera shows the potential for noninvasive diagnosis of disease and severity. The radiation delivered to the tissue is diffused due to the heterogeneity of the biological structures and absorption by the chromophores as it propagates through the tissues. The characteristics of the absorption and scattering of rays in the tissue change as the disease progresses, therefore the reflectance analysis shows quantitative diagnostic information on tissue pathology. ¹⁰

The hyperspectral camera is used to diagnose and differentiate skin cancers. Cancer cell metabolism differs molecularly from healthy skin; thus, the distribution of chromophores is different, for example, malignant melanoma. This affects the amount of radiation absorbed and the reflectance of the tissue. Aloupogianni at el. ¹¹ indicate that the distribution of chromophores in the tissue may be like a fingerprint, and that a different arrangement and concentration of chromophores may suggest a disease state.

The aim of the study was to assess the hyperspectral reflectance of acne skin before and after IPL therapy and to compare it with the reflectance of healthy skin.

2. MATERIALS AND METHODS

The study involved 27 volunteers aged 24 ± 5 years, including 18 women and nine men diagnosed with moderate acne, abnormal keratinization, excessive seborrhea, discoloration, and ductal opening of sebaceous glands around the nose, forehead, chin, and cheeks (group A). The control group (C) consisted of 20 people (12 women and eight men, mean age 25 ± 3 years) without acne lesions. The study obtained a positive opinion from the Ethics Committee of the SUM Number: PCN/0022/KB1/11/I/20, and the volunteers provided their voluntary written consent to participate in the study. All acne volunteers underwent a series of four treatments using IPL Lumecca (InMode MD Ltd., Yokneam, Israel) at weekly intervals. A head emitting a wavelength of 515–1200 nm was used for the treatments, and the light energy was increased with each treatment. The following were used in subsequent treatments (I‐IV): I—energy 12 J and a short pulse, II—energy 13 J or 14 J depending on the skin reaction after the first treatment and a short pulse, III ‐ 14 J and a short pulse and 11 J and long pulse, IV ‐ 15 J and short pulse and 11 J and long pulse.

The volunteers were asked to clean their face with a cleansing gel 3 h before the test and not to use any other cosmetic preparations. This was to simulate the physiological condition of the skin and to avoid light waves being reflected or absorbed by active substances in cosmetics. In the study, the right and left profiles were photographed. The body and face of the volunteers were set in the same position because it was used a chair and a standardized mat from the FotoMedicus system Elfo, Poland, where the photos are taken in specific positions to ensure the identity and repeatability of the photos.

The volunteers with acne lesions were photographed with the Specim IQ hyperspectral camera (Specim, Oulu, Finland) before starting the series of treatments (A1) and 1 week after the fourth treatment (A2). The imaging of the skin of healthy people in the control group (C) was performed once. The images have a static size of 512 × 512 pixels, with 204 bands ranging from 397 to 1004 nm and a spectral resolution of 3 nm. This enables the use of a movable prism (spectrograph), which splits the radiation beam and then projects it onto the light‐sensitive element. Wavelengths from 400 to 950 nm were used for the research due to measurement errors and artifacts at the beginning and end of the spectral spectrum. According to the manufacturer, the visible area is 0.55 × 0.55 m, which allows for a spatial resolution of 1.07 mm at a distance of 1 m from the object. The image preprocessing includes only image calibration against the calibration panel. To obtain results that can be compared between individual volunteers, it is necessary to calibrate the recorded data. It is done automatically by the camera software. Each scene must be recorded with a calibration panel with a known level of reflectance (18% in this case). The camera software allows to indicate on the image a calibration panel against which all other pixels are calibrated.

The hyperspectral imaging (HSI) is an innovative technology for obtaining quantitative measurements based on transcutaneous spatial and spectral information. The images obtained from a hyperspectral camera can be used to evaluate reflectance values in order ​​to determine tissue properties. An additional advantage is that HSI is noninvasive and does not require special preparation of the patient for the examination, apart from maintaining the physiological conditions of the skin. Images corresponding to a specific wavelength were isolated from the hyperspectral image of acne skin, the ROI (region of interest) areas were arbitrarily determined, and they were extracted from each hyperspectral image.

2.1. Statistical analysis

The statistical analysis of the results was performed using Statistica 13 software. The significance of changes in skin reflectance was assessed using the Friedman ANOVA test and Dunn's post hoc test.

3. RESULTS

The more acne lesions are visible in the image, the lower the reflectance of skin with lesions. Figure 1 shows the difference in the appearance of acne skin before and after IPL treatment and skin without acne lesions. The A1 photography shows lofty and reddened acne lesions. The posttreatment inflammatory lesions healed leaving discoloration, and the skin was devoid of swell (Figure 1—photo A2). The skin shows different reflectance for different wavelengths. The brighter the image and the less noticeable changes are made, the higher the reflectance value (Figure 2). For some wavelengths, acne changes are clearer than for others, due to the presence of chromophores in the skin. The hemoglobin and melanin have their own characteristic light absorption peaks or intervals.

FIGURE 1.

RGB (red‐green‐blue) photos taken with a hyperspectral camera: acne skin before treatments (A1) and after intense pulsed light (IPL) therapy (A2), and skin without acne lesions (control—C).

FIGURE 2.

The acne skin reflectance at particular wavelengths before treatment (A1) as a sample. The images have been converted to grayscale.

The chart (Figure 3) shows a lower reflectance for acne skin before IPL treatment ranging from 400 to 600 nm. For acne skin after treatments and skin without acne changes (C), the reflectance chart coincides in the initial section. From 500 nm, points slightly diverge and acne skin shows lower reflectance after the therapy. At the point of 430 nm, these graphs coincide, which means that the acne skin after the treatments shows the same reflectance as the skin not affected by acne lesions. The IPL therapy increases skin reflectance, which may be due to fewer chromophores in the skin, less inflammation, and fewer acne lesions.

FIGURE 3.

The average skin reflectance of the ROIs for three groups of patients before treatment (A1), after treatment (A2) and control (C), for the spectral range of 400–950 nm. (A) The average skin reflectance values from 400–600 nm in magnification. (B) The average skin reflectance for the spectral range of 400–950 nm.

There are three lowest points on the graph of acne skin reflectance before the treatment. These are the wavelengths: 430, 549, and 588 nm. The reflectance of the examined areas was compared to those points determined by the wavelengths (Figure 4).

FIGURE 4.

The acne skin reflectance before (A1) and after treatments (A2) with the use of intense pulsed light (IPL) and the skin of a healthy control group (C) measured at 430, 549, and 588 nm. *p < 0.05.

No statistical significance was found between the reflectance of acne‐prone skin before and after the treatments at the wavelength of 430 nm (p = 0.273). The difference between the reflectance of acne skin before the treatments and the control was statistically significant (p = 0.028), and no significant difference was shown between the reflectance of the acne skin after the treatments and the control (p = 0.273).

A statistically significant difference was found between the reflectance of acne skin before the treatments and the reflectance of the skin after a series of IPL treatments at the wavelengths of 549 (p = 0.028) and 588 nm (p < 0.001).

There is statistical significance between before IPL treatments and control values at the wavelengths of 549 (p < 0.001) and 588 nm (p < 0.001). The statistical significance was demonstrated between the reflectance of acne skin after treatment and the control for the wavelengths of 549 (p = 0.028) and no significance for 588 nm (p = 0.1).

The values of acne skin reflectance after the treatments are higher than before the therapy, but they are significantly lower than the control values. The skin unaffected by acne lesions has a higher reflectance in the studied spectral range.

4. DISCUSSION

The HSI—an optical imaging technology that combines a camera with a spectrometer—provides both spatial and spectral information of the analyzed structures. The tissue/light interaction (reflection, absorption, scattering) generates specific spectral signatures, enabling a noninvasive and nonionizing qualitative and quantitative analysis of the object's biochemical composition: the so‐called optical biopsy. ¹² The HSI enables the rapid differentiation of materials and is therefore used in remote sensing and vegetation control ¹³ or forensics. ¹⁴ The HSI allows to distinguish pathological tissues from healthy ones; therefore, this technology has been developed in medicine in recent years. ¹⁰ It is a noninvasive, painless, quick method that does not require any special preparation of a patient.

Thanks to HSI, it is possible to determine the wavelength that is reflected by the skin to the smallest extent, and thus is most absorbed by the tissue. This will enable a precise selection of the wavelength, for example, of a laser, or the limitation of the IPL range with cut‐off filters in the desired range. This will help to minimize side effects such as erythema, burns, or low effectiveness due to the optimization of the treatment procedure in terms of skin reflectance. Following this lead, the most effective laser would be to penetrate to the appropriate depth and show the lowest reflectance.

In our study, we used the IPL device for acne therapy due to numerous literature reports on the effectiveness of this method in reducing acne lesions. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ The effectiveness of the therapy was assessed with the use of hyperspectral imaging, which showed similar values of acne skin reflectance after treatment and healthy control skin. This may suggest a change in the amount of chromophores in the skin, which affects the reflectance value.

The low reflectance in the range of 400–599 nm may be caused by the accumulation of chromophores in the skin that absorbs waves in this range, ²⁰ which would explain the lowest skin reflectance in cases with moderate acne. It is associated with the current inflammation, also with the participation of melanocytes (which increases the likelihood of postinflammatory hyperpigmentation), increased microcirculation (acne lesions are characterized by lymphocytic infiltration and an increase in the expression of inflammatory vascular markers, e.g., VCAM‐1, ICAM‐1), which promote angiogenesis ²¹ , ²² and involving Cutibacterium acnes, producing porphyrins. ²³

The melanin absorbs a wide range of the light spectrum, and the longer the wavelength, the less light is absorbed by melanin. ²⁴ The melanin is one of the first chromophores to encounter light in its path. Its main task is to protect tissues from the harmful effects of ultraviolet radiation. ²⁵ The melanocytes are also involved in the ongoing inflammation, which stimulates them to produce melanin. The inflammation in the course of acne often disappears with postinflammatory discoloration. ²⁶ As a result of inflammation, the blood supply to the tissue increases and it is supplied with blood, which is the main absorber of radiation due to the presence of hemoglobin. In circulation, there are derivatives of hemoglobin: oxyhemoglobin, deoxyhemoglobin, and methemoglobin. The absorption of light energy by oxyhemoglobin and deoxyhemoglobin peaks in the range 400–420 nm and 540–577 nm, gradually decreasing at longer wavelengths. The deoxyhemoglobin has an absorption maximum of 420 and 580 nm. Oxidized hemoglobin (oxyhemoglobin) shows the highest absorption at 410 nm and has two secondary peaks in the range of 550−600 nm. The absorption intensity of hemoglobin directly depends on its amount, so the more hemoglobin in the tissue, the more it absorbs radiation. For wavelengths from 500 to 600 nm, the maximum absorption of melanin and hemoglobin occurs. ²⁷ , ²⁸ , ²⁹

An interesting phenomenon is also the abrupt increase in reflectance above 600 nm, which remains at the reflectance value of 0.8–0.9. Light beam scattering is assumed to dominate absorption. This assumption is less important for skin below 600 nm due to the high absorption rates of hemoglobin and melanin. ²⁴ In the case of soft tissues, the spectral range for which the light radiation penetrates the deepest is defined by the therapeutic or optical window and is in the range 650–1200 nm, that is, in the red and near‐infrared ranges. In this range, radiation penetrates the skin most deeply, and its absorption is the lowest. The depth of penetration of the beam of the laser world or IPL depends both on the properties of the tissue, and also on the wavelength, energy density, and treatment spot. ³⁰

5. CONCLUSIONS

1. Acne skin properties of absorbing light radiation are different than those of healthy skin.

2. Intensive pulsed light therapy increases the reflectance of acne‐prone skin, which coincides with healthy skin, which may be due to a reduction in the content of chromophores in the skin, less inflammation, and fewer acne lesions.

3. The use of a noninvasive method of hyperspectral imaging allows for monitoring of the course of the disease and the effectiveness of therapy at every stage. It also enables the precise selection of the light wavelength for the therapeutic treatment, and consequently increases the effectiveness of the treatments, enhancing their safety, and optimizing the number of irradiations.

4. Additionally—due to the more and more frequent antibiotic resistance of bacteria—acne therapy with intense pulsating light seems to be a good alternative to antibiotic therapy.

CONFLICT OF INTEREST STATEMENT

The authors have no conflict of interest to declare.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.