Results in Physics 11 (2018) 975–983

Contents lists available at ScienceDirect

Results in Physics
journal homepage: www.elsevier.com/locate/rinp

Light source optimizing in a biphotonic vein finder device: Experimental and T

theoretical analysis
A. Madrid García , P. R. Horche
⁎

Departamento de Tecnología Fotónica y Bioingeniería, Universidad Politécnica de Madrid, Madrid, Spain

ARTICLE INFO ABSTRACT

Keywords: Solutions oriented to non-invasive superficial veins visualization by physicians are not numerous neither cheap.
Vein finder In spite of the advantages these solutions can provide in cannulation or venepuncture processes, especially in
Biophotonics some population groups, their use is not widespread. Making use of an own designed 3D printed model, we
LED experimentally studied the feasibility of using Ultra-Bright light-emitting diodes (LEDs) of different optical
Biomedical device
features to facilitate the visualization of the veins in a transillumination biophotonic approach.
Skin optics
Cannulation
Intravenous therapy
Blood extraction
Transillumination
3D printing
MATLAB

Introduction pollicization [5] or sclerotherapy [6].

For the reasons given above, we study different optical configura-
According to some studies, around eighty per cent of the hospita- tions in a biophotonic device, to maximize veins visualization and to
lized patients receive a peripheral intravenous (I.V) cannulation [1]. In assess the potential benefits of using these devices in the healthcare
fact, although it is considered as a routine procedure, their related assessment.
complications are often unknown and undervalued. Some associated
medical complications, (which also have attached other complications State of the art
such as economical indemnities) derived from unsuccessful cannulation
procedures, can be bacterial infection, extravasation, phlebitis, throm- Nowadays, there are different approaches by which a physician can
bosis, embolism or nerve damage [2]. see in a non-invasive way the patient’s vessel. A summary of all of them
A published study in the British Journal of Anaesthesia showed the is presented in Fig. 2.
ratio of difficult venepuncture cases in different populations group. The
percentage of those is shown in Fig. 1. 1. Ultrasound (US) biomedical imaging systems have been used for
The amount of adipose tissue surrounding vessels, the loss of muscle assisting cannulation from a long time ago. The non-invasive tech-
tone, a high melanin concentration, a blood flow reduction or an in- nique, the real-time acquisition and the portable features have done
jured skin are some cases which can prevent a successful venepuncture from US one of the most common ways for I.V. procedure. Site-Rite
process at the first attempt. Prevue® Ultrasound System, is an example of US systems already
Furthermore, there are some cases in which choosing the most used.
suitable vein is even more important than doing the cannulation or 2. Inside biophotonic based solutions there are two main alternatives:
venepuncture process accurately: avoiding sensitive places derived 2.1. Solutions that take advantage of Near Infrared (NIR) spectro-
from venous fibrosis, local infections or inflammations, fistulae or scopy principles: these solutions project over the skin surface a
vascular grafts, and patients with haemophilia [4] are some examples of map of the blood vasculature employing a visible wavelength.
that situations. Aside from venepuncture, a vein visualization system The projections are achieved by previously radiating the skin
could also abe useful in some other medical procedures, like surface with NIR light and quantifying the amount of light

⁎
Corresponding author.
E-mail addresses: [email protected] (A. Madrid García), [email protected] (P. R. Horche).

https://doi.org/10.1016/j.rinp.2018.10.033
Received 11 August 2018; Received in revised form 23 October 2018; Accepted 23 October 2018
Available online 03 November 2018
2211-3797/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 3. Light propagation in the skin schema.

Fig. 1. Percentage of cannulation difficult cases [3].
acquiring reflectance information thanks to an RGB conventional
absorbed by the blood (haemoglobin) and reflected by sur- camera and multispectral Wiener estimation [7].
rounding tissues. The device will capture the information, 4. Finally, other approaches not fully developed, are being under re-
process the data and finally project over the skin surface the search such as a penetration method using NIR [8], robotic systems
venous distribution. AccuVein®, VeinViewer® and IV-eye®, are or pressure sensors to facilitate blood extraction [9].
the most well-known devices which take advantage of this
principle. A fully description and comparison among the approaches pre-
2.2. Solutions that take advantage of visible light together with skin sented in Fig. 2, can be found in [10]. Although, ultrasound systems
optics properties, also known as transillumination solutions: have been demonstrated to be useful [11], they are not as affordable as
they are based on the use of LED technology. Skin surface is other solutions based on biophotonics which also facilitate the cannu-
irradiated with a light source of an appropriate visible wave- lation procedure [12,13]. Performance differences between NIR and
length, allowing directly the visualization of the veins by dif- transillumination solutions have also been discussed in recent years
ferences in the biological and physical tissues properties. [14,15].
Absorption, scattering and reflexion of the light inside skin Nowadays, transillumination methods are becoming more popular
tissue will determine the visualization since not all wavelengths since they present two major benefits over those based on NIR princi-
are able to penetrate enough into the skin. Therefore, the ples: portability and cost. However, available transillumination com-
emitting wavelength in this approach is consciously chosen to mercial solutions differ significantly among them. The number of light
maximize the light penetration. Absorption differences between sources employed, from two light-emitting diodes up to thirty-two, the
skin chromophores and blood chromophores will allow veins wavelength, or the luminous intensity are some of the optical para-
visualization. Venoscope®, Veinlite®, Wee-Sight meters which can vary from one commercial solution to another.
Transilluminator® and Illumivein® solutions follow this ap- In this study, we identified and analysed the light source parameters
proach. The intensity of the light reflected is usually higher that can affect the veins visualization, since there is not an agreement of
than the light transmitted to the vessels. Therefore, difficulties the optimal light source features to be employed in these biomedical
of superficial vein visualization arise, given that too powerful devices. Results derived from this study can be useful to develop an
light and shadows patterns could emerge. To overcome this accurate vein seeker device or to understand and choose wisely from
issue, side-transillumination method can be used instead of current transillumination solutions.
transillumination, which uniformly illuminates a small region
of the skin to reduce the shadows and allowing light penetra- Light transport in the skin
tion up to 6 mm in depth.
3. Smartphones have also been used in the vein visualization task by Skin tissue structure is designed to reduce light penetration through
a successive layer structure with different composition and properties

Fig. 2. Different non-invasive vein visualization approaches.

976
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 4. Light propagation schema in RTT theory.

Table 1 Approximation which assumes that light is able to penetrate deeper into
Design parameters. the tissue, since absorption is lower than scattering [21,22].
Critical Parameters
However, physical responses derived from a tissue after having been
irradiated with an incident light beam not only depends on the tissue
Wavelength ( ) : distance between consecutive crests on a periodic wave of the light biological properties but also on the features of the light source. Factors
emitted by the LED diode. Only wavelengths inside the visible range, such as the wavelength of the incident light, the optical power emitted
380–750 nm, will be considered
by the light source, the exposure rate (W/cm2), the irradiated area, the
Luminosity (IV ) : luminous power per unit solid angle emitted by the LED diode in a
irradiation time, the number of pulses and their duration and the po-
particular direction. The studied luminosity range in this study varies from 150
mcd to 15.000 mcd larization state will influence that tissue response. The exposure size of
Angle ( ) : angle at which the light emitted by the LED diode is distributed. Angles the light also affects the effective penetration depth, achieving a deeper
between 15° and 140° were studied penetration when the intensity given by the source of light is greater. L.
Other Parameters Fodor et al. carried out a description of some of these factors in [23].
Therefore, in medical applications or devices which take advantage
Number of light sources (N): number of LEDs used in each device. Two different of the interactions between light and matter (i.e. biophotonic applica-
values were considered: 11 and 20
tions), like the vein finder, the properties of the light source should not
Polychromatic light (PL): devices with more than one different LED and wavelength
will be considered as polychromatic light devices be considered in an isolate way, but also tissue properties and physical
Size of the light source (S): size of the LED diode lens. Two possible values were optical phenomena should be taken into consideration at the same time.
considered: 3 mm and 5 mm

Experimental set up
[16,17]. Differences in these optical, physical and biological properties
Design parameters
explain why vessels can be seen when the skin is irradiated with visible
light of an appropriate wavelength. Inside optical and physical phe-
To determine the most suitable light-source and structural features
nomena, reflection, scattering and absorption plays a key role. Whereas
in a biophotonic vein finder approach device that optimize the visua-
the first one, reflection, limit the light propagation at the first light-skin
lization, we considered and ranked different parameters that could af-
encounter, absorption restricts light penetration and scattering deviates
fect vein visualization in a biophotonic approach, Table 1. From top to
the light trajectory inside the skin [18]. Therefore, it is estimated that
bottom, and according to our experiments, we established the most
about the 4%–7% of the light is reflected as a consequence of refractive
decisive physical parameters that could be involved in the visualization.
index differences between air and skin. Considering that the longer the
Critical parameters are those parameters that must be satisfied to vi-
wavelength of the light, the deeper the penetration [19], as it is re-
sualize vessels, otherwise visualization will not be achieved. Other
presented in Fig. 3, the transmitted visible light will not penetrate
parameters are those than can modify the visualization but are not
deeper than 4–5 mm inside the skin.
mandatory to visualize vessels.
In human tissues, scattering is the dominant factor affecting light
propagation. Inside skin, keratin and melanin from the epidermis form
the main scattering barrier. In dermis layer, where blood vessels are Implemented devices
located, scattering is produced by heterogeneities derived from the
variable cellular components presented in the different sublayers (e.g. Different devices were built to compare they visualization perfor-
stratum corneum or basale) and by different molecular concentrations. mance depending on the different parameters’ settings. Table 2 sum-
Due to the great number of scatterers presented in the media -obstacles marizes the different devices as well as their features. To identify them,
that deviate the light path from a straight trajectory-, mathematical we built their name according the next schema:
models like the Radiation Transport Theory (RTT) have been developed
W(avelength) (n- X (Colour name’s initial: N (umber YY S (ize of Y
to describe light propagation when multiple scattering phenomena are
m) Orange, Red, Combined) of LEDs) LEDs) (mm)
presented. This theory is based on tracking photon changes (energy
Case design of WON11S5, WRN11S5, WCN11S5 devices were retrieved
flow) in an infinitesimal volume dV , within differential solid angle d
from [24]. In addition, by using a 3D printing approach, with polylactic
around s direction, Fig. 4. Changes in energy flow can be caused by
acid as the chosen structural material, the rest of the devices were built.
incoming, outgoing, absorbed and emitted photons, that have been
Ergonomic and mobile features were considered; therefore, a
scattered from one scatterer (e.g. elastin or collagen fibres) to the light
minimalist, small and portable device was designed. Energy supply is
from other scatterers existing in the media. Energy flow implies energy
provided by an accumulator and light sources are distributed along the
conservation, so, RTT can be interpreted as an energy conservation
ring to achieve a uniform and homogeneous illumination inside the
problem. A further description of this theory can be found in [20].
FOV (27 mm). Both structural and electrical schemas are shown in
The study of RTT in human skin is done by using the Diffusion
Fig. 5.

977
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Table 2
Datasheet of the devices.
Visual design Name λ (nm) IV (mcd) α° N PL S (mm)

WON11S5 600–610 4.200–5.800 15 11 N 5

WRN11S5 620–630 4.900–6.300 35 11 N 5

600–610 4.200–5.800 15 5 5
Y
WCN11S5
620–630 4.900–6.300 35 6 6

WON20S3 600–610 7.000–8.000 30 20 N 3

WRN20S3 650–670 5.800–7.000 30 20 N 3

600–610 7.000–8.000 30 10 3
Y
WCN20S3
650–670 5.800–7.000 30 10 3

WGN20S3 520–530 8.400–10.000 30 20 N 3

Fig. 5. Design process of the devices.

Chosen light sources decision was based on the theoretical pene- penetrate enough into the skin to reach blood vessels and to be reflected
tration power of the different wavelengths, as was seen in Fig. 3, and again to the surface. Therefore, the most decisive factor of a veins vi-
also on a light source experimental comparison. In Fig. 6, all the dif- sualization system, in a transillumination approach, is the wavelength
ferent tested and evaluated light sources can be found: of the light employed. An evaluation of the different light sources with
different wavelengths was carried out in order to assess skin absorption
Wavelength and veins visualization. We tested light sources with a wavelength
Veins visualization can only be achieved by irradiating the skin with range from 465 to 640 nm. Only light sources above a wavelength
a light of an appropriate wavelength, otherwise light would not threshold that we set in 580 nm (Fig. 6. 3b), deep enough in the skin

978
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

and reach blood vessels making the visualization possible. Upper limit
would be in the frontier of visible to infrared light, that is, 760 nm.
Orange and red colours fall inside the range of these wavelengths,
however, the results between both of them differ. Whereas with red
wavelengths (620–670 nm, Fig. 6. 5a, 5b, 5c) the illumination reaches
further horizontal areas, facilitating to follow the veins direction, with
orange ones (590–610 nm, Fig. 6. 4a, 4b) a better contrast is achieved
and therefore more structures are visible, Fig. 7.
WGN20S3 device was implemented to show that although there
exists an absorption peak in 590 nm, green light cannot penetrate the
epidermis layer, hence, no visualization is achieved. It is important to
highlight that WGN20S3 is the device with the highest luminous in-
tensity (10.000 mcd). From this point, we discarded devices and light
sources with a wavelength under 580 nm.

Luminosity
Only LED’s with enough luminous intensity (commercially known as
Ultra-bright LEDs) are able to achieve a homogeneous illumination
inside the field of view (FOV) of the device. The FOV must be com-
pletely illuminated, hence, larger FOV will require higher luminous
intensity. Without a uniformly illumination vessels will not be correctly
distinguished, so that, luminosity is a critical parameter in a transillu-
mination veins visualization system.
We tested LED diodes in a range of 150 to 15.000 mcd, those with
enough intensity can be seen in Fig. 6, 1-5 a, b. Considering the FOV of
our designed device (27 mm) good visualization results appear from
4.000 mcd, so only, Ultra-bright LEDs were used. However, in some
cases, good visualization is achieved with fewer millicandelas (Fig. 6,
5c). Furthermore, increasing the intensity over a value, in our case
7.000 mcd, does not seem to be related with better results.

Angle
The light source angle must ensure a uniform illumination. LED
diodes with a very small angle < 15° will illuminate isolated areas, in a
spot way, creating black holes in the centre of the FOV and avoiding the
visualization. However, wide angles (Fig. 6. 6a, 6b) do not focus enough
the light. Fig. 8 shows the importance of a good angle selection.
Whereas the left device, WON11S5, angle is 15° and the light pattern
follows an isolate distribution, the right prototype illumination dis-
tribution is homogeneous due to its higher angle (30°), WCN20S3.
Common commercial LED angle range is 30°–60°, which is perfectly
Fig. 6. Light source study. 1a) IV :12.000–15.000 mcd, :30°, : 465–475 nm. valid for this kind of applications.
1b) IV : 1.600–4.200 mcd, :15°, : 465–475 nm. 1c) IV : 330–500 mcd, :30°, :
465–475 nm. 2a) IV : 8.400–10.000 mcd, :30°, : 520–530 nm. 2b) IV : Results
7.000–15.000, :15°, : 518–524 nm. 2c) IV : 150–330 mcd, : 30°, : 570 nm.
3a) IV : 5.800–7.000 mcd, : 30°, : 585–595 nm. 3b) IV : 5.500–9.000 mcd,
Results are split into two categories depending on the study pur-
:15°, : 588–594 nm. 3c) IV : 120–150 mcd, : 30°, : 585–595 nm. 4a) IV :
7.000–8.000 mcd, :30°, : 600–610 nm. 4b) IV : 4.200–5.800 mcd, : 15°, :
pose:
600–610 nm. 5a) IV : 5.800–7.000 mcd, :30°, : 650–670 nm. 5b) IV :
2.200–4.200 mcd, : 15°, : 620–625 nm. 5c) IV : 500–700 mcd, :30°, : 1) To evaluate the light source parameters, under the best light con-
640 nm. 6a) IV : 330–500 mcd, :120°, : 585–595 nm. 6b) IV : 220–330 mcd, ditions, in a dark environment.
:140°, : 600–610 nm. 2) To evaluate the performance of the devices under real conditions, in
a well illuminated environment, under different skin tones in

Fig. 7. Green, orange and red forearm illumination differences with devices: WGN20S3, WON20S3, WRN20S3. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)

979
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 8. Importance of the light source angle. WON11S5, WRN11S5 and WCN20S3 devices.

Fig. 9. Left forearm of a young woman.

Fig. 10. Left forearm of a young man.

volunteers with different ethnic background. sources, polychromatic lights and size of the light source were studied.
To begin with, the required number of light sources depends directly on
the length of the FOV and inversely to the luminous intensity and the
Evaluation of the light source parameters angle. We have not seen practical differences by increasing the LEDs
number from 11 (WON11S5, WRN11S5, WCN11S5) to 20 (WON20S3,
Pictures from different anatomical vessels were taken and processed WRN20S3, WCN20S3), Fig. 9.
to show the performance of each device under ideal conditions, in a Also, the LEDs size (3 mm vs 5 mm) does not seem to take part in a
dark environment. Together with wavelength, luminosity and angle,- better visualization. As happened with the number of light sources,
parameters related to the device structure such as the number of light

980
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 11. Left forearm of a young woman.

diodes must ensure a uniform light distribution, no matter the size. independently of the device employed under light conditions, and can
LED’s front was not studied in this work. be tracked along the whole arm. No visualization differences are ap-
When studying the benefits of using polychromatic light sources, preciated in these Mediterranean volunteers of different gender, age
this is, using in the same device LEDs with different wavelengths, we and weight.
appreciate that results do not differ or improve the results achieved The main difference of individuals in the second group is the age.
with monochromatic devices. For instance, in Fig. 10, device WCN11S5 Blood vessels are harder to be found on V. 3 in well-light environments
visualization is akin to WON11S5 and WCN20S3 is very similar to than on V. 4. In fact, only a small portion of them are seen and cannot
WON20S3 and WRN20S3 devices. be followed along the whole arm. As occurred with V.1 and V. 2, the
The light sources distribution, this is, the sequence of consecutive performance of orange devices is greater than the performance of red
diodes with different colour, does not seem to be involved in a better devices. WCN20S3 device is more effective than WRN11S5, WCN11S5
visualization: For instance, in Fig. 11, device WCN11S5 sequence is 1:1 and WRN20S3. On the other hand, vessels from V. 4, an elderly man,
(red-orange), while the sequence of WCN20S3 device is arbitrary. No are appreciated with all devices although there exist differences be-
differences at a glance are found. tween red pure devices and orange ones. Again, WON20S3 evidences
Experimentally, better visualization results are achieved with better results.
WON20S3, the orange device with more diodes, due to its better con- Differences between a normal weight and an obese individual are
trast Figs. 9, 10. However, in Fig. 11 the results among all devices are considered in the third group. Only the most superficial area of the
almost the same. Isolate and big vessels are found with all of them, vessels is visualized when applying devices with eleven diodes to V. 5.
however, little and tortuous structures are better found with wave- However, WON20S3, WRN20S3 and WCN20S3 devices shows correctly
lengths among 590–610 nm. the vessels of the wrist. No greater differences are found between
An image processing analysis based on histogram features was WON20S3, and WRN20S3, as occurred with other volunteers. On its
carried out under MATLAB®. Images in Fig. 10 were firstly turned to a behalf, visualization results obtained in V. 6 are similar to those ob-
grayscale intensity image. Afterwards, a contrast-limited adaptive his- tained in the first group. Devices with twenty LEDs perform better, and
togram equalization and filtering processes were done. Fig. 12 shows superficial veins can be tracked in the whole forearm.
the result. Devices do not work on the Asian volunteer, V.7 as well as on the
Histogram images taken from the different vein finder devices are other volunteers. Although, device WON20S3 gives a glimpse of small
similar, with a homogeneous peak distribution. However, as we con- vessels portions, the major part of these remain hidden. No practical
cluded before, some structures are visually enhanced when orange differences are found between WRN11S5 and WRN20S3, and between
colour is presented in the vein finder device, both in an isolate way, as WCN11S5 and WCN20S3. Conversely, blood vessels from the dark skin
also combined with other wavelengths. volunteer, V. 8, are easily found with all devices. In this case, luminous
intensity from the incident light source is the most critical parameter
(rather than the wavelength), since vessel’s contour are perfectly
Performance of the devices in real conditions
identified when using devices with twenty diodes.

Since medical procedures are usually done in well-light environ-

ment, the performance of the devices was also studied in those condi- Conclusions
tions, Fig. 13.
Analyses were run under a total of eight volunteers (n = 8). Table 3 Outcome images from the different evaluated devices, with different
outlines the different demographical and physiological volunteers’ (V.) wavelengths, are very similar among them in ideal conditions. It seems
features: that there is a better-contrast trend in tortuous vein structures when
Volunteers were split into groups of two individuals to evaluate orange light (600–610 nm) is employed. Other optical parameters such
different demographical and physiological features. as the angle or the luminous intensity will be involved in the device
Pure orange devices (WON11S5 and WON20S3) provide the best performance visualization regardless the employed wavelength. Based
results in group 1. Moreover, devices which use more diodes on our results, any device with a wavelength between 585 and 670 nm,
(WON20S3, WRN20S3 and WCN20S3) are better than those with less an angle between 15 and 35° and a luminous intensity of 4000 mcd (or
diodes (WON11S5, WRN11S5 and WCN11S5). Veins are easily found, higher) which covers the whole FOV will ensure a better vein

981
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 12. Image processing results with their normalized histograms.

visualization. understand and design a biophotonic vein finder device, that can sup-
The performance of the devices in real conditions vary among vo- port the physician in the healthcare practice. It has been proven that
lunteers. Overall, WON20S3 is the best device for visualizing vessels, these devices can be meaningful in different tasks like the non-invasive
regardless the skin tone of the volunteer. In dark skin tones it seems that superficial veins visualization and also for obtaining a map of the veins,
the most determinant parameter is the luminous intensity, rather than which, after an image processing, can be useful in the diagnosis of
the wavelength. In very fair skin tones, like the Asian individual and the certain pathologies.
child, we found more difficulties to find large vessel sections.
Differences based on other aspects like the gender do not seem to be
involved in a different visualization among individuals. However, since Acknowledgement
we have studied a limited sample, further studies increasing the size of
the group are required. This work was partly supported by the Spanish MINECO under
To conclude, results derived from this study can be useful to Grant TEC2015-65212-C3-2-P and TEC2017-84065-C3-1-R.

982
A. Madrid García, P. R. Horche Results in Physics 11 (2018) 975–983

Fig. 13. Performance of WCN20S3 and WON11S5 in well-light environment in a young Mediterranean volunteer.

Table 3 visualizing device. Sensors 2017;17(304):1–19. https://doi.org/10.3390/

Volunteers demographical and physiological features. s17020304.
[9] Balter ML, Chen AI, Fromholtz A, Gorshkov A, Maguire TJ, Yarmush ML. System
Volunteer Race Gender Age range Weight Group design and development of a robotic device for automated venipuncture and di-
agnostic blood cell analysis. 2016 IEEE/RSJ international conference on intelligent
1 Mediterranean Male 23 Normal 1 robots and systems (IROS). IEEE; 2016. p. 514–20.
2 Mediterranean Female 56 Overweight [10] Cantor-Peled G, Halak M, Ovadia-Blechman Z. Peripheral vein locating techniques.
3 Mediterranean Male 13 Normal 2 Imaging Med 2016;8(3).
4 Mediterranean Male 87 Normal [11] van Loon FHJ, Buise MP, Claassen JJF, Dierick-van Daele ATM, Bouwman ARA.
Comparison of ultrasound guidance with palpation and direct visualisation for
5 South American Female 45 Class I obesity 3
peripheral vein cannulation in adult patients: a systematic review and meta-ana-
Indian
lysis. Br J Anaesth 2018.
6 South American Male 23 Normal
[12] Barreras J, Chang TP. Using a near infrared device to improve successful venous
Indian access in children with special health care needs. J Assoc Vascular Access
7 Southeast Asian Female 24 Normal 4 2017;22(2):75–80.
8 African Male 24 Normal [13] Hosokawa K, Kato H, Kishi C, Kato Y, Shime N. Transillumination by light-emitting
diode facilitates peripheral venous cannulations in infants and small children. Acta
Anaesthesiol Scand 2010;54(8):957–61.
[14] Heinrichs J, Fritze Z, Klassen T, Curtis S. A systematic review and meta-analysis of
References
new interventions for peripheral intravenous cannulation of children. Pediatr
Emerg Care 2013;29(7):858–66.
[1] Rivera AM, Strauss KW, van Zundert AAJ, Mortier EP. Matching the peripheral [15] Haile D, Suominen PK. Technologies in pediatric vascular access: have we improved
intravenous catheter to the individual patient. Acta Anæsthesiologica Belgica success rate in peripheral vein cannulation? Acta Anaesthesiol Scand
2006;58(1):19. 2017;61(7):710–3.
[2] Quincy Tran. Ultrasound vs Veinviewer in Patients With Difficulty IV Access. [16] Ross MH, Pawlina W. Histology: a text and atlas: with correlated cell and molecular
https://clinicaltrials.gov/ct2/show/NCT02618252. Last Update Posted, January biology. 6th ed.; 2011.
23, 2018. [17] Ash C, Dubec M, Donne K, Bashford T. Effect of wavelength and beam width on
[3] Chiao FB, Resta-Flarer F, Lesser J, Ng J, Ganz A, Pino-Luey D, et al. Vein visuali- penetration in light-tissue interaction using computational methods. Lasers Med Sci
zation: patient characteristic factors and efficacy of a new infrared vein finder 2017;32(8):1909–18.
technology. Br J Anaesth 2013:aet003. [18] Ding H, Lu JQ, Wooden WA, Kragel PJ, Hu XH. Refractive indices of human skin
[4] Guillon P, Makhloufi M, Baillie S, Roucoulet C, Dolimier E, Masquelier AM. tissues at eight wavelengths and estimated dispersion relations between 300 and
Prospective evaluation of venous access difficulty and a near-infrared vein vi- 1600 nm. Phys Med Biol 2006;51(6):1479.
sualizer at four French haemophilia treatment centres. Haemophilia [19] Lister T, Wright PA, Chappell PH. Optical properties of human skin. J Biomed Opt
2015;21(1):21–6. 2012;17(9):090901.
[5] Ficke BW, Ransom EF, Oakes JE. Near-infrared vein visualization in index finger [20] Tsia Kevin K. Understanding biophotonics: fundamentals, advances, and applica-
pollicization. Surgical Tech. 2017;42(6):481.e1–2. tions. Pan Stanford; 2015. p. 90.
[6] Zharov VP, Ferguson S, Eidt JF, Howard PC, Fink LM, Waner M. Infrared imaging of [21] Keiser G. Biophotonics: concepts to applications. Springer; 2016.
subcutaneous veins. Lasers Surg Med: The Official J Am Soc Laser Med Surg [22] Baranoski GV, Krishnaswamy A. Light and skin interactions: simulations for com-
2004;34(1):56–61. puter graphics applications. Morgan Kaufmann; 2010.
[7] Song JH, Kim C, Yoo Y. Vein visualization using a smart phone with multispectral [23] Fodor L et al. Aesthetic applications of intense pulsed light, 11 DOI: 10.1007/978-1-
Wiener estimation for point-of-care applications. IEEE J Biomed Health Inf 84996-456-2_2, © Springer-Verlag London Limited 2011.
2015;19(2):773–8. [24] Instructables, A. (2017, September 28). 3d Printed Medical Vein Finder. Retrieved
[8] Kim D, Kim Y, Yoon S, Lee D. Preliminary study for designing a novel vein- from http://www.instructables.com/id/3d-Printed-Medical-Vein-Finder.

983