Objective Measurement of Hypertrophic Scars using Skin

Colorimeter

Iveta Bryjova

, Jan Kubicek

, Vladimir Kasik

, Daniel Kamensky

, Hana Klosova

, Marek Penhaker

and Martin Cerny

VSB–Technical University of Ostrava, FEECS, K450, 17. Listopadu 15, Ostrava-Poruba, Czech Republic

Burn Centre, University Hospital Ostrava, 17. Listopadu 1790 Ostrava-Poruba, Czech Republic

Keywords: Skin Colorimeter Burns, Scars, Prototype.

Abstract: The paper deals with the methodology of the scars pigmentation objective assessment and their time

evaluation on the base acquired data with the skin colorimeter prototype DSC1 (Detection of Scar Color).

The analysis is primarily focused on the hypertrophic scars pigmentation assessment after heeling of deep

burns which often exhibit the pigmentation. In the process of the scars evaluation in some patients it goes to

the spontaneous pigmentation changes. If the pigmentation changes long-term persist and patient requires

corrections, various treatment methods can influence these pigmentation changes (for instance the laser

therapy and others). In the context of the complex development evaluation and in the process of the scars

treatment, these changes are commonly observable well but their quantification is usually difficult, therefore

using of the objective methods is desirable. The particular kind of such objective method is the skin

colorimeter. The technical concept and testing of the skin colorimeter prototype DSC1 is presented in this

paper.

1 INTRODUCTION

Hypertrophic scarring of burns represents the most

frequent complication of the trauma, especially in

the deep burns when both epidermis and dermis are

destroyed. In the case of the normal circumstances,

the healing process goes in three phases:

inflammation, proliferation and remodelling phase.

In the case of the complicated healing, the risk of

developing hypertrophic, functionally restrictive and

aesthetically objectionable scars is particularly high.

(Blazek et al., 2015), (Cerny et al., 2008)

Hypertrophic scars develop within the primary

wound and protrude over the level of the

surrounding skin, they are painful, tough, itching,

initially red and may progress to scarry contractures

requiring surgical treatment. Deep burn scars often

present also pigmentation disorders in the terms of

decreased pigmentation, the so-called

hypopigmentation, increased pigmentation, the so-

called hyperpigmentation or a mix of different

intensity thereof. The pigmentation disorders

affecting the scar make it more visible against the

healthy tissue which is aesthetically less acceptable

for the patient. Pigmentation disorders exhibit from

multiple factors and have not been clarified exactly

yet. The main indications of the burns are the depth

and scope of the burn, activity and duration of the

inflammatory phase of healing, also various cellular

mediators (NO, histamine) and other internal tissue

factors which influence melanogenesis in

melanocytes. At the same time certain exogenous

influences enter the game, such as UV radiation. The

factors mentioned above are often antagonistic,

therefore, the resulting changes in the pigmentation

are difficult to predict and highly individual. There

are several scales for clinical evaulation: Patient and

Observer Scar Assessment Scale, Visual Analog

Scale, Manchester Scar Scale and the most

frequently used Vancouver Scar Scale which uses

pigmentation classification, elevation, pliability and

vascularisation of the scar. Scar assessment using

the above scales is inherently subjective – it depends

upon the experience and skills of the physician

performing clinical assessment. (Augustynek et al.,

2010), (Scafide et al., 2016), (Shin et al., 2015),

(Stekelenburg et al., 2016)

126

Bryjova I., Kubicek J., Kasik V., Kamensky D., Klosova H., Penhaker M. and Cerny M.

Objective Measurement of Hypertrophic Scars using Skin Colorimeter.

DOI: 10.5220/0006147101260133

In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 126-133

ISBN: 978-989-758-216-5

2 STATE OF ART

Subjective methods for the evaluation of the color

changes of hypertrophic scarring after burn trauma

are widely described in scientific articles published

in high-impact journals worldwide. Despite this fact,

a objective method for burns assessment is still

missing. The recent research shows that there are not

many published scientific papers that would

specifically focus on the objective evaluation of

color hypertrophic scarring after burn injury.

Published results of clinical trials in particular

describe the effect of UV radiation linked with an

increasing incidence of skin cancer. In paper

(Klosová et al., 2013) the authors publish the results

of clinical measurements carried out on 27 male and

31 female probands, in age ranging from 6 to 9

years. The main objective of the study was to

demonstrate that the incidence of skin cancer in

adulthood is closely linked to the action of

ultraviolet radiation in childhood. For an objective

assessment of the skin color a commercial

measuring device – namely, colorimeter Chroma

Meter CS-200 (Konica Minolta, Japan) – was used.

The results proved very high accuracy of the

measurement.

In (Štětinský et al., 2015) the authors publish the

results of a comparative study of two different

methods of objective assessment of depigmentation

using reference colorimetric methods. The

measurement was performed by the colorimeter

Chroma Meter CS-200 (Konica Minolta, Japan) and

digital camera with polarization spectroscopic

technology TiVi600 (Tissue Viability Imager

TiVi600, WheelsBridge AB, Linkoping, Sweden).

The results prove the fact that TiVi600 non-contact

sensing achieves more accurate results than

colorimeter Chroma Meter CS-200.

Another comparative study was performed with

the target of evaulation the potential of selected

parameters measurements (measurement accuracy,

sensitivity and reproducibility) of a new commercial

device Antero 3D (Miravex Limited, Ireland) with

leading commercial dermatology devices

Mexameter MX 18 (Courage Khazaka, Germany)

and Colorimeter CL 400 (Courage Khazaka,

Germany). The results of comparative analysis of

these devices showed that Antero exhibits more

sensitivity melanin and also improved resolution

capability of erythema and melanin. The sensitivity

of Mexameter and Colorimeter is almost identical.

Based on the results of the search of available

commercial equipment for the detection of human

skin color, or melanin concentration, we proceeded

to the actual realization of the prototype skin

colorimeter DSC1 order to objectify color change

hypertrophic scars after thermal trauma in clinical

practice. The main focus was to design a device that

will be compatible and safe for clinical use, but also

economically feasible.(Cerny et al., 2009),

(Kukucka, 2009), (Machaj et al., 2016), (Romanelli

et al., 2013), (Verhaegen et al., 2014).

3 HARDWARE DESIGN OF SKIN

COLORIMETER

The functional unit of the colorimeter DSC1 is

composed from several commercially available

electronic components: color sensor TSC230,

programming board Arduino Uno with

microprocessor ATMega328 and alphanumeric LCD

display 16x2. For the initial test run wiring

breadboard with the Arduino kit connection cables

were applied. In the first step, validation of the

measuring using low-cost optical components was

performed. The individual hardware parts are

represented in the block diagram (fig. 1). The

essential component of the device is the

programmable color sensor TSC230. The sensor is

able to detect any number of colours, and works on

the principle of light intensity into frequency

transformation. This part include 4 illuminating LED

diodes in squared configuration, and the so-called

RGB field (fig. 2). (Penhaker et al., 2013)

The RGB field is placed in the middle of the

sensor and contains 64 quartz photodiodes. Each

photodiode is equipped by 3 filters intended for the

detection of red, green and blue color. 16

photodiodes are not equipped by filters, and they are

primarily used for the scanning an detection of the

white illumination. (Penhaker et al., 2011)

Digital input and output of the sensor is

facilitated by the communication interphase with the

microcontroller AVR ATMega 328 which is built in

the programming board Arduino Uno. The sensor is

linked with the microcontroller by six digital I/O

lines which allows for picking the respective color

(R/G/B), sensor sensitivity (Power down /2 % /20 %

/100 %), output instruction and one pin for signal

output. In the output we obtain rectangular signal

(fig. 3), its frequency (AA) is given by the

photodiode current. (Majernik et al., 2014), (Marek

and Krejcar, 2015)

Objective Measurement of Hypertrophic Scars using Skin Colorimeter

127

Figure 1: The block diagram of the DSC1 hardware part.

Figure 2: XY RGB photodiode field of the TSC230

sensor.

Figure 3: Principle of rectangular signal transformation.

The frequency f

represents the sensor detection

function and is described by the following formula:

).(Re

Eeff

(1)

where f

denotes the output frequency, f

denotes the

output frequency for darkness state, i.e. when ܧ݁ = 0

as a result leakage currents, ܧ݁(ܹ݉ ܿ݉

) is the

intensity of the radiation incidence and ܴ݁(݇ܪݖ

(ܹ݉/ܿ݉

)) denotes the sensor reaction to the

wavelength of the respective light. Due to the fact

that frequency f

is directly proportional to

brightness of the individual colour components, it is

possible to represent the appropriate output

frequency in RGB color model and obtain the

resulting colour. During the start-up it is possible to

calibrate two levels in the RGB space – absolute

black color is represented by zero coordinates given:

[0, 0, 0] which then represents the darkness status of

constant, and absolute white color is given: [255,

255, 255] which denotes the maximum RGB level

also called the white balance. These levels therefore

define the brightness scale of the individual

components of the RGB model [0 - 255].

(Augustynek and Penhaker, 2011)

After plugging-in of the sensor (fig.4) the power

supply and communication with the programming

board was tested. A bypassing capacitor (reduction

of high frequency current flow) and a LCD display

for simple viewing of the data measured were

connected to the feeder circuits. We use the

potentiometer (10 ݇Ω) linked with the LCD display

to adjust the required contrast. There is also push-

button switch for activation of the circuit. (Penhaker

et al., 2012), (Vybiral et al., 2011)

Figure 4: Testing HW connection of the sensor with the

programming board.

The last step in completion of DSC1 was the design

(fig. 5, 6, 7, 8) and 3D print of the protective cover

to meet the hygiene requirements for the clinical use.

Figure 5: The protective cover – the below view.

Figure 6: The protective cover – the back view.

BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices

128

Figure 7: The protective cover – the front view.

Figure 8: The protective cover – the right view.

(Kaartinen et al., 2011), (Kubicek et al., 2016),

(Lammers et al., 2011).

4 SOFTWARE DESIGN OF SKIN

COLORIMETER

The associated software of colorimeter prototype is

designed in the JAVA language in the Arduino IDE

environment, and it utilizes functionalities of the

ElecFreaks library, especially functionalities for

signal frequency measurement and LCD display

control.

The controlling algorithm is described in the

flow chart (fig.9). In the first step, the LiquidCrystal

lcd() functionality and the #define clause were used

to define communication pins of the LCD display

and the sensor. In the consecutive step, the sensor is

initialized and the frequency scale is adjusted on 2 %

(enables measurement with higher sensitivity).

Consequnetly, the sensor is calibrated to level [255,

255, 255] which in RGB space corresponds with the

absolute white color. By this way, the upper limit of

the RGB space is established. Calibration sensor is

done by mat white plate or cardboard. Calibration

must be done during the first start or restart.

(Bryjova et al., 2016).

Figure 9: Algorithm flow chart of the DSC1 skin

colorimeter.

5 CLINICAL TESTING AND

RESULTS

Verification of the reliability and the accuracy of the

device are tested on the base clinical measurements;

just a few selected cases are discussed. Case 1

represents a male aged 28 with a hypertrophic scar

caused by a thermic trauma (fig.10) at the dorsal

side of the right arm. Altogether 10 measurements of

the hypertrophic scars and the same number of

control measurements of the healthy arm are

performed.

Objective Measurement of Hypertrophic Scars using Skin Colorimeter

129

Figure 10: Arm affected by the hypertrophic scar.

The results of the measurements are summarized

in the table (tab. 1) and on the scatter plot (fig. 12)

for each color component R, G, B; the scatter plot

moreover contains luminance calculated from color

components according to the standard formula Y =

0.299R + 0.587G + 0.114B. Values representing the

scars are plotted by triangles, values representing the

control measurements are plotted by the circles, the

individual color components and luminance are

presented in corresponding colors and in black. The

results are further represented by the box plot (fig.

11) showing the measurements of the scars (see the

left part of the diagram) and control measurements

of the healthy skin (see the right part of the

diagram). These graphic outputs are supplemented

by comparison of the color and luminance of the

hypertrophic scar and healthy skin.

Table 1: The overview of the RGB values measured for

the scary and healthy parts of the arm skin.

Diversity of color components and luminance is

statistically tested in the form of null hypothesis of

equation of mean values of dependent data samples,

sequentially for the color components and

luminance. For testing the Location Test

functionality implemented in the Mathematica

software version 10.4 developed by Wolfram

Research Inc. was applied which automatically

selects the optimum way of testing based on pre-

tests of the data samples. In this particular case, the

pair Student´s t-test is selected for the red and green

components and luminance and the signed rank test

was used for the blue component.

Figure 11: Box Plot – the arm affected by the hypertrophic

scar.

The results are unambiguous – all the RGB

components and luminance Y in the scary skin at the

level of significance 0.05 are statistically

significantly lower (scarry skin is darker as shown in

the above comparison of colors, p-values for R, G, B

and Y components being sequential 8 x 10-7, 5.2 x

10-5, 0.015 and 6.8 x 10-6.

Figure 12: Scatter Plot – arm affected by the hypertrophic

scar.

Case 2 is a patient hospitalized in the Burns Centre

of the Teaching Hospital in Ostrava (fig. 13). The

testing involved measurement of extensive mature

hypertrophic scars following an injury by the electric

current. The patient agreed with the measurement

and signed an informed consent. The measurement

is done under constant ambient conditions:

temperature 22.8 °C, relative humidity 30.7 %,

BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices

130

patient´s body temperature 36.2 °C. Both the scarry

and healthy parts of the body are tested separately 10

times. Two anatomical locations are measured

(shoulder and crus).

Figure 13: Affected shoulder after electrical current injury.

For the statistical imaging and testing analogical

procedure is used as in the case 1. In this particular

case, the pair Student´s t-test is selected for all color

components as well as luminance. The results are

again unambiguous – all the RGB components as

well as luminance Y in the scarry skin are

statistically significantly lower (scarry skin is darker

as shown in the above comparison of colors, box

plot (fig. 14) and scatter plot (fig. 15), p-values for

R, G, B and Y components being sequentially 2.2 x

-9

, 1.0 x 10

-8

, 3.0 x 10

-10

and 7.5 x 10

-11

Table 2: Overview of the RGB values measured for the

scary and healthy parts of the shoulder skin.

Figure 14: Box Plot – shoulder after the electrical current

injury.

Figure 15: The Scatter Plot – the shoulder after the

electrical current injury.

Left crus

Statistical imaging and testing is performed using

analogical procedure as in the case 1. In this

particular case, the pair Student ´s test is selected for

red and green components and luminance and the

signed rank test is used for the blue component. R

and G color components as well as luminance Y in

the scarry skin are statistically significantly lower

(scarry skin is darker as shown in the above

comparison of colors, box plot (fig. 17) and scatter

plot (fig. 18), p-values for R, G and Y being

sequentially 0.001, 0.019 and 0.012

Figure 16: The crus with the hypertrophic scar.

Objective Measurement of Hypertrophic Scars using Skin Colorimeter

131

On the contrary, in the blue component we do

not reject the null hypothesis of equation of the

mean value on the basis of p-value 0.437. Here,

difference in luminance is less obvious though

noticeable even when observing and comparing

colors by naked eye.

Table 3: Overview of the RGB values measured for the

scary and healthy parts of the crus skin.

Figure 17: Box Plot – the crus with the hypertrophic scar.

Figure 18: Scatter Plot – the crus with the hypertrophic

scar.

6 CONCLUSIONS

Testing results prove that areas containing scars are

statistically significantly darker than the normal

skin, and it all them color components, including the

luminance. There is a one exception. Just in the case

of the left crus the blue scar component is

insignificantly darker. The proposed prototype of the

skin colorimeter DSC1 seems to be promising

device in the context of the objective assessment of

the color skin weaker differences (hypertrophic and

other scars) in the clinical conditions.

The device primarily allows for the

quantification of the pigmentation level of the

hypertrophic scars as the consequence after

extensive thermally traumas. In the context of the

future clinical measurements the next physical and

other conditions will be determined for valid values

measurement which consequently will be able to

represent and assess the correlation with the most

frequently used clinical scale VSS (Vencouver Scar

Scale).

The next steps should be focused to the

miniaturization and improving of the hardware

components of the DSC1. These improvements are

necessary to routine using of the DSC1 for the

clinical measurement as affordable, reliable and

accurate method.

ACKNOWLEDGEMENT

This article has been supported by financial support of TA

ČR, PRE SEED Fund of VSB-Technical univerzity of

Ostrava/TG01010137. The work and the contributions

were supported by the project SV4506631/2101

'Biomedicínské inženýrské systémy XII'.

REFERENCES

Augustynek, M., Labza, Z., Penhaker, M., Korpas, D., &

Society, I. C. (2010). Verification of set up dual-

chamber pacemaker electrical parameters. 2010

Second International Conference on Computer

Engineering and Applications: Iccea 2010,

Proceedings, Vol 2, 168-172.

doi:10.1109/iccea.2010.187.

Augustynek, M., & Penhaker, M. (2011). Non invasive

measurement and visualizations of blood pressure.

Elektronika Ir Elektrotechnika(10), 55-58.

doi:10.5755/j01.eee.116.10.880.

Blazek, P., Krenek, J., Kuca, K., Krejcar, O., Jun, D., &

Ieee. (2015). The biomedical data collecting system.

2015 25th International Conference Radioelektronika

(Radioelektronika), 419-422.

Bryjova, I., Kubicek, J., Dembowski, M., Kodaj, M.,

Penhaker, M. 2016. Reconstruction of 4D CTA brain

perfusion images using transformation methods,

Advances in Intelligent Systems and Computing, 423,

pp. 203-211.

BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices

132

Cerny, M., Martinak, L., Penhaker, M., & Rosulek, M.

(2008). Design and implementation of textile sensors

for biotelemetry applications. In A. Katashev, Y.

Dekhtyar, & J. Spigulis (Eds.), 14th nordic-baltic

conference on biomedical engineering and medical

physics (Vol. 20, pp. 194-197).

Cerny, M., & Penhaker, M. (2009). Circadian rhythm

monitoring in homecare systems. In C. T. Lim & J. C.

H. Goh (Eds.), 13th international conference on

biomedical engineering, vols 1-3 (Vol. 23, pp. 950-

953).

Kaartinen, I.S., Välisuo, P.O., Bochko, V., Alander, J.T.,

Kuokkanen, H.O. 2011. How to assess scar

hypertrophy - A comparison of subjective scales and

Spectrocutometry: A new objective method. Wound

Repair and Regeneration, 19 (3), pp. 316-323.

Klosová, H., Štětinský, J., Bryjová, I., Hledík, S., and

Klein, L. 2013. Objective evaluation of the effect of

autologous platelet concentrate on post-operative

scarring in deep burns. Burns.,s. -. DOI:

10.1016/j.burns.2013.01.020.

Kubicek, J., Bryjova, I., Penhaker, M. 2016. Macular

lesions extraction using active appearance method

Lecture Notes of the Institute for Computer Sciences,

Social-Informatics and Telecommunications

Engineering, LNICST, 165, pp. 438-447.

Kubicek, J., Bryjova, I., Penhaker, M., Kodaj, M.,

Augustynek, M. 2016. Extraction of myocardial

fibrosis using iterative active shape method Lecture

Notes in Computer Science (including subseries

Lecture Notes in Artificial Intelligence and Lecture

Notes in Bioinformatics), 9621, pp. 698-707.

Kukucka, M. (2009). Modeling of logic diagnostic system

knowledge base evaluation.

Lammers, G., Verhaegen, P.D.H.M., Ulrich, M.M.W.,

Schalkwijk, J., Middelkoop, E., Weiland, D., Nillesen,

S.T.M., Van Kuppevelt, T.H., Daamen, W.F. 2011. An

overview of methods for the in vivo evaluation of

tissue-engineered skin constructs Tissue Engineering -

Part B: Reviews, 17 (1), pp. 33-55.

Machaj, J., Brida, P., & Benikovsky, J. (2016). Scalability

optimization of seamless positioning service. Mobile

Information Systems. doi:10.1155/2016/9714080

Majernik, J., Jarcuska, P., & Ieee. (2014). Web-based

delivery of medical education contents used to

facilitate learning of infectology subjects. 2014 10th

International Conference on Digital Technologies (Dt),

225-229.

Marek, T., & Krejcar, O. (2015). Optimization of 3d

rendering in mobile devices. In M. Younas, I. Awan, &

M. Mecella (Eds.), Mobile web and intelligent

information systems (Vol. 9228, pp. 37-48).

Penhaker, M., Darebnikova, M., & Cerny, M. (2011).

Sensor network for measurement and analysis on

medical devices quality control. In J. J. Yonazi, E.

Sedoyeka, E. Ariwa, & E. ElQawasmeh (Eds.), E-

technologies and networks for development (Vol. 171,

pp. 182-196).

Penhaker, M., Klimes, P., Pindor, J., & Korpas, D. (2012).

Advanced intracardial biosignal processing. In A.

Cortesi, N. Chaki, K. Saeed, & S. Wierzchon (Eds.),

Computer information systems and industrial

management (Vol. 7564, pp. 215-223).

Penhaker, M., Kasik, V., & Snasel, V. (2013). Biomedical

distributed signal processing and analysis. In K.

Saeed, R. Chaki, A. Cortesi, & S. Wierzchon (Eds.),

Computer information systems and industrial

management, cisim 2013 (Vol. 8104, pp. 88-95).

Scafide, K.N., Sheridan, D.J., Taylor, L.A., Hayat, M.J.

2016. Reliability of tristimuluscolourimetry in the

assessment of cutaneous bruise colour Injury, 47 (6),

pp. 1258-1263.

Shin, J.U., Kang, S.-W., Jeong, J.J., Nam, K.-H., Chung,

W.Y., Lee, J.H. 2015. Effect of recombinant human

epidermal growth factor on cutaneous scar quality in

thyroidectomy patients Journal of Dermatological

Treatment, 26 (2), pp. 159-164.

Stekelenburg, C.M., Hiddingh, J.,Kuipers, H.C.,

Middelkoop, E.,Nieuwenhuis, M.K., Polinder, S., Van

Baar, M.E. 2016. Cost-effectiveness of laser doppler

imaging in burn care in The Netherlands: A

randomized controlled trial Plastic and

Reconstructive Surgery, 137 (1), pp. 166e-176e.

Štětinský, J., Klosová, H., Kolářová, H., Šalounová, D.,

Bryjová, I. and Hledík, S. 2015. The time factor in the

LDI (Laser Doppler Imaging) diagnosis of burns.

Lasers in Surgery and Medicine. 47(2): 196-202. DOI:

10.1002/lsm.22291. ISBN 10.1002/lsm.22291. ISSN

01968092.

Romanelli, M., Dini, V., Mani, R. Skin and vascular

assessments 2013. Measurements in Wound Healing:

Science and Practice, pp. 193-223.

Kubicek, J., Penhaker, M., Bryjova, I., Augustynek, M.

2016. Classification method for macular lesions using

fuzzy thresholding method) IFMBE Proceedings, 57,

pp. 239-244.

Verhaegen, P.D.H.M., Bloemen, M.C.T., Van Der Wal,

M.B.A., Vloemans, A.F.P.M., Tempelman, F.R.H.,

Beerthuizen, G.I.J.M., Van Zuijlen, P.P.M. 2014. Skin

stretching for primary closure of acute burn wounds

Burns, 40 (8), pp. 1727-1737.

Vybiral, D., Augustynek, M., & Penhaker, M. (2011).

Devices for position detection. Journal of

Vibroengineering, 13(3), 531-535.

Objective Measurement of Hypertrophic Scars using Skin Colorimeter

133