Contact Dermatitis 2002, 46, 129–140
Printed in Denmark . All rights reserved
Copyright C Blackwell Munksgaard 2002
Guidelines for visualization of cutaneous blood flow
by laser Doppler perfusion imaging
A report from the Standardization Group of the European Society of Contact
Dermatitis* based upon the HIRELADO European community project
A. F1, M. S2, K.-P. W3, K. W4, C. A5, T. F6, G. E. N4
1Department of Dermatological Research, Leo Pharmaceutical Products Ltd, Ballerup, DK-2750 Denmark,
2Dermatological Department of Ruhr Universitet Bochum, D-44791 Germany,3proDERM Institute for Applied
Derm. Res. GmbH, Schenenfeld/Hamburg, D-22869 Germany,4Department of Biomedical Engineering,
Linköping University, SE-58185 Linköping, Sweden,5Department of Dermatology, University Hospital, S-581 85
Linköping, Sweden,6National Institute of Working Life, S-171 84 Solna, Sweden
This report reviews how to set up a laser Doppler perfusion imaging system intended for visualiza-
tion of skin blood perfusion, capture images and evaluate the results obtained. A brief summary
of related papers published in the literature within the areas of skin irritant and allergy patch
testing, microdialysis and skin tumour circulation is presented, as well as early applications within
other fields such as diabetology, wound healing and microvascular research.
microcirculation; measurement guidelines. C Blackwell Munksgaard, 2002.
Accepted for publication 8 January 2002
bioengineering methods; cutaneous blood flow; laser Doppler perfusion imaging;
Laser Doppler measurement of cutaneous blood
flow was introduced more than 2 decades ago fol-
lowing the development of laser technologies and
fibre-optic systems (1–3). Various devices operat-
ing with a Helium-Neon laser as light source and
a fibre-optic cable fixed in contact with the skin
surface have came into use. Blood flow is measured
under a single spot of skin defined by the incident
and Doppler-shifted reflected light. The recorded
in situ sample is sized about 1mm (3), the exact
dimensions being dependent on the tissue optical
properties and the probe geometry (4). This
sample mainly includes the superficial vascular
*Members of the standardization group on non-invasive
(Rome), A. Bircher (Basle), P.J. Coenraads (Groningen),
T. Fischer (Stockholm), P.J. Frosch (Dortmund), J.-M.
Lachapelle (Brussels), A. Lahti (Oulu), H.I. Maibach
(San Francisco), J. Roed-Petersen (Copenhagen), J. Ser-
up, chairman (Copenhagen), R.A. Tupker (Nieuwegein),
P. van der Valk (Nijmeegen), J.E. Wahlberg (Stockholm).
plexus. Relative perfusion values representing the
total blood flow in capillaries and small sized ves-
sels are recorded and the results are expressed in
arbitrary units. The local site variation in blood
flow recorded with such single spot devices is
major, and the number of measurements required
is determined by the desired precision in the cu-
taneous blood flow estimate (5). This is a logical
consequence of the microanatomical resolution of
single spot laser Doppler flowmetry.
(LDPI) was introduced with systematic X–Y scan-
ning of the laser beam and horizontal mapping of
the cutaneous blood flow in a larger area of the
skin (6–9). This emerging medical technology is a
non-contact system with the laser beam position
controlled in the X and Y directions by a system
of mirrors. A digital image composed of numerous
single-point recordings forms a two-dimensional
flow map of the blood flow variability over an ex-
tended skin surface. This image is displayed on a
computer monitor and can be stored, analysed,
and printed out.
Cutaneous blood flow is a dynamically fluctuat-
ing biological variable, which possesses substantial
130FULLERTON ET AL.
spatial heterogeneity, and it was soon understood
that many parameters influence its magnitude and
regulation. The European Society of Contact Der-
matitis in 1994 published formal guidelines on
laser Doppler flowmetry based upon experiences
with the early single-point systems (10). These
guidelines summarized individual-related, environ-
mental-related and technique-related variables and
established the arterial occlusion/reactive hyper-
aemia experiment as an aid to assess instrumental
performance, especially the biological zero level
and the dynamic range. Additionally, the Euro-
pean Society of Contact Dermatitis in 1996 pub-
lished guidelines on colourimetric measurement of
skin colour according to spectrophotometric and
tri-stimulus principles (11). Blood flow and colour
(redness) are different albeit overlapping manifes-
tations of vascular functions.
The present guidelines on LDPI resulted from a
multicentre European Union project named HIR-
ELADO (HIgh REsolution LAser DOppler per-
fusion imaging in dermatology, project number IN
10178I) involving centres in Stockholm, Linköp-
ing, Copenhagen, Hamburg and Bochum. The
guidelines address mainly technical aspects of
LDPI, standard usage, application principles and
interpretation of the images generated. Individual-
related and environment-related variables are cov-
ered by the previous guidelines (10). The HIREL-
ADO group used the Perimed/Lisca PIM Laser
Doppler Perfusion Imager in this project. This
technology was originally developed by two of the
authors (Nilsson and Wårdell), who founded the
company Lisca in the early 1990s. The present
guidelines are, however, aimed to be generic and
relevant also for LDPI instrumentation produced
by other manufacturers.
In addition to the laser Doppler technique,
many other methods for the assessment of tissue
blood perfusion have been presented in the scien-
tific literature. Of those methods, orthogonal po-
larization imaging (12), video-photometric capilla-
roscopy (13) and thermographic imaging (14),
constitute technologies that create images directly
reflecting or relating to the tissue blood perfusion.
The two former technologies, however, create im-
ages that include only a small number of micro-
vessels representing the local blood perfusion,
while thermographic imaging techniques visualize
the tissue surface temperature rather than the dy-
namics of the underlying blood perfusion.
Cutaneous blood flow
The cutaneous blood flow is mainly in the service
of thermoregulation and an instrument for main-
tenance of a constant body core temperature under
the influence of different environmental con-
ditions. Its net function is the controlled release of
metabolic heat. This is mainly achieved through
adjustments of vascular tone and skin perfusion.
Peripheral vessels show vasomotion with rhythmic
contractions (6–8/min), which may be important
for vessel-to–tissue interactions and exchange of
nutrients, gas and heat. Vascular tone is dependent
on control by the autonomic nervous system, on
cerebral functions including emotional stimuli,
ocular stimuli and sounds, and on a variety of
other factors such as orthostatic position relative
to the heart level and gravity, nutritients, medi-
cines, smoking, etc. The cutaneous blood flow can,
under physiological conditions, be reduced to less
than half of its basal resting value but increase 20-
fold to reach maximal capacity. The vasculature
is normally in a relatively constricted state. Thus,
measurement of cutaneous blood flow requires
understanding and control of many variables other
than the instrumental factors.
The local blood perfusion of the skin may be
increased in various skin disorders associated with
inflammation and reduced in conditions with vaso-
constriction and occlusion.
The light of a laser Doppler flowmeter pen-
etrates the skin variably to a depth of about 0.6
mm (15). The gradually decreasing sensitivity with
depth in the dermis allows the sampling volume to
be probed in a non-uniform manner, with an aver-
age measurement depth of about 0.3mm (16). The
vasculature within the sampling volume consists of
vessels of variable size, function, tension, tonicity
and microanatomical orientation relative to the in-
cident and reflected laser light (Fig.1).
Rete vessels and vessels around adnexae are
special and differ from the superficial plexus. Re-
gional differences are major and the head, palms
and soles represent hyperperfused zones. Due to
the in-sample and between-sample heterogeneity in
biological structure and function of the microvas-
culature, it is ambiguous to express the cutaneous
blood flow in absolute terms, for example as
ml¡gª1¡minª1tissue. Instead, emphasis should be
put on careful and detailed description of experi-
mental conditions, instrument validation and set-
ting, selection and preconditioning of study sub-
jects, the measurement site selected, environment
and laboratory room control and experimental de-
sign. Of further importance are adequately se-
lected and explained comparative situations, etc.,
in order to demonstrate significant relative flow
changes and regional flow distributions, as results
of experiments which may be repeated and the re-
sults reproduced at another study centre using an-
other LDPI device. In practice, LDPI technology
is well suited for comparisons of the average cu-
131LASER DOPPLER IMAGING
Fig.1. Schematic diagram of the superficial cutaneous micro-
vasculature and the photon pathways in tissue.
taneous blood flow recorded at two separate skin
areas within the same image. Furthermore, pro-
vocative experiments may be employed where the
resting-state blood flow before a stimulus can be
used as a reference for the increase or reduction in
flow instituted by the stimulus. In either case, the
multitude of measurement points made available
by the use of LDPI (as opposed to single-point
laser Doppler flowmetry) overcomes the problem
with cutaneous blood flow heterogeneity, and
allows for adequate statistical testing of possible
differences in blood flow between selected skin
areas within the same image or at comparative ex-
LDPI is a non-invasive technology that creates
two-dimensional flow-maps of tissue perfusion
with no need for physical contact with the tissue
or use of dyes and tracer elements. The technology
was first commercialized in the early 1990s. In one
embodiment of LDPI technology (7), a laser beam
scans the tissue step-wise by means of a computer-
controlled mirror system housed in a camera-like
scanner head, which is positioned about 10–30cm
above the tissue surface (Fig.2). This brand of
LDPI is commercially available from Perimed/Lis-
Fig.2. Split drawing of the LDPI scanner head showing laser,
mirrors, beam path, object, scanning pattern and detector.
ca AB, Sweden. Two versions (standard and high
resolution) of the PIM laser Doppler perfusion im-
ager were used by partners in the HIREALDO
project to evaluate the generic performances of
this medical technology. Other commercial ver-
sions of LDPI technology utilize a continuously
moving laser beam (8) and different ways of pro-
cessing the Doppler signal (9).
In the LDPI technology used by the HIRELA-
DO group, a low power (1mW) laser beam, gener-
ated by a solid-state laser operating at the wave-
length 670nm, is directed through a mirror system
to the skin, where it forms a light spot with a size
of about 1mm2. As the beam is successively moved
in discrete steps over the skin surface, the back-
scattered light is collected by a photo-detector po-
sitioned inside the scanner head. At each measure-
ment site the beam is arrested for a short period of
time, during which the back-scattered light, partly
Doppler-shifted by the moving blood cells in the
superficial microvascular network, is detected. On
the surface of the detector, frequency-shifted and
non-frequency-shifted light is mixed to form a
photo-current. The frequency content of the fluc-
tuating portion of this photo-current is related to
the blood cell average speed, while the magnitude
is determined primarily by the number of moving
blood cells within the scattering volume. After pro-
cessing, a signal that scales linearly with tissue per-
fusion is formed that is proportional to the prod-
uct of the blood cell speed and concentration. This
signal is stored in the computer memory and the
laser beam is automatically moved to the next
measurement site, where the procedure is repeated.
When all measurement sites have been sampled,
the stored values are displayed as a colour-coded
perfusion image in parallel with a black-and-white
photo image of the object. This perfusion image
can be further analysed by the integrated system
software as well as exported to other packages.
132FULLERTON ET AL.
Fig. 3. Image examples.
By use of the system setting software, images can
be recorded one at a time or as a sequence of im-
ages. The size and resolution of the images as well
as the speed by which they can be captured are set
separately by the user. The individual images can
be displayed in Relative or User defined colour
scale. Using Relative colour scale, the colour scale
is stretched between the smallest and the largest
perfusion value. This setting is useful for visualiz-
ing the spatial heterogeneity in tissue perfusion
within a single image. By use of the User defined
colour scale, the user sets the end-point perfusion
values of the colour scale. This setting is useful
when spatial differences in low perfusion areas are
to be visualized. In addition, a black-and-white
photo of the object based on the back-scattered
laser intensity can be generated. Various examples
of perfusion images are shown in Fig.3.
When the LDPI system is set to Monitor mode,
which enables the recording of tissue perfusion in
the time domain, the laser beam can be moved to
a certain site on the tissue surface where it will
record the tissue perfusion continuously in a single
or small number of adjacent sites. The result is dis-
played as a time trace on the computer monitor.
The possibility of probing the tissue perfusion at
adjacent sites in a cyclic pattern enables the time
traces to be displayed as average perfusion values.
This feature yields a higher reproducibility at re-
peated measurements and suppresses the influence
of variation in tissue perfusion at adjacent sites,
which is one of the main limitations of fibre-optics-
based laser Doppler perfusion monitors (Fig.4).
By use of the Imaging and the Monitor mode,
both the spatial heterogeneity and the temporal
variability of tissue perfusion can be investigated.
The Image mode is most useful when perfusion in
different tissue areas in an image is to be com-
pared, while Monitor mode is to be employed in a
provocative experiment in which tissue perfusion is
133LASER DOPPLER IMAGING
Fig.3. Image examples.
134FULLERTON ET AL.
Fig.4. A time trace of cutaneous blood flow showing the effect of occlusion (left) and venous stasis (right) on the recorded signal.
expected to change over time. If both temporal and
spatial variability are of concern, the LDPI system
should be set-up to record a series of images in
Laser Doppler perfusion imaging systems process
the Doppler signal in a way similar to that of laser
Doppler perfusion monitors. The output signal is
linearly related to the product of the blood cell
speed and concentration within the scattering vol-
ume. Like other equipment with a linear response,
a two-point calibration procedure is sufficient. As
the ‘no-flow’ reference point the zero value is gen-
erally selected. This ‘no-flow’ reference point value
is obtained by calculating the average value of an
image created by use of a diffusely scattering solid
material, such as a piece of white cardboard or a
flat, white plastic disk. The ‘high-flow’ reference
point can best be generated by imaging an object
with known, stable and uniform internal motion.
By use of standardized calibration boxes and
uniformly selected target values for the ‘high-flow’
reference point, the variation in interinstrumental
sensitivity and intrainstrumental drift can be kept
to a minimum. Standardization of the calibration
boxes is, however, a complex task because their re-
producibility, long-term stability and precision
need to be significantly better than the instrument
that is to be calibrated. The currently available
calibration methods for laser Doppler equipment,
based on recording of the mobility of microspheres
in random motion, therefore need further improve-
ment and evaluation.
The possibility of maintaining a low variability
of results obtained by different users (interuser
variability) is mainly dependent on careful design
of the individual experiment, control of environ-
mental factors and test subject/patient selection, as
well as instrumental set-up. If these factors are
carefully considered and established protocols are
followed, the interuser variability can generally be
kept low, because the image capturing process is
inherently user-independent. For the same reasons,
the interlaboratory variability can be kept low pro-
vided that the precautions are carefully considered.
Preparing for the capturing of perfusion images in-
cludes assessment of both environmental and sub-
ject-related factors. The main factors and variables
that influence cutaneous blood perfusion are (from
major to minor impact): temperature, anatomical
site, physical and mental activity, food and drug
intake, temporal (day-to-day), temporal (diurnal),
posture, age, sex, menstrual cycle and race. Since
these aspects have been dealt with in previous
guidelines (10), this section will focus on how to
set up the system and choose the adequate image
parameters for best system performance.
Marking and identifying areas of interest
In order to be able to identify an area of interest
in the individual images, small ink dots or tape of
a colour that effectively absorbs the laser light may
be placed as markers on the skin surface coincid-
ing with, for example, the corners of the area to
be investigated. Since these identifiers tend to ab-
sorb the laser light they will show up against the
135LASER DOPPLER IMAGING
background colour in the perfusion image and the
associated photo image. The size of such markers
should be no smaller than about 1mm2. The
markers should be applied at least 15min before
image capturing starts in order to avoid uninten-
tional local alteration in skin perfusion. It is fur-
thermore important that the skin sites to be inves-
tigated are left uncovered during the acclimatizing
period prior to image capturing, because of the in-
fluence of the ambient temperature on the tissue
Ambient light level
The ambient light level must be kept to a minimum
in order to avoid interference between ambient
light and the laser beam. Although images can
readily be recorded under different light con-
ditions, the influence of ambient light may alter the
magnitude of the actual perfusion values to a de-
gree dependent on the light source employed (light
bulbs or fluorescent light) and its intensity. Inter-
ference from ambient lighting frequently manifests
itself as a regular speckled pattern appearing in the
photo image. The user is encouraged to record test
images of the skin site under study with the light
on and off and to calculate the average values of
identical areas of interest, in order to find out
whether the influence of ambient light can be toler-
ated in the actual situation.
Distance to tissue
The distance between the lower edge of the scanner
head and the skin can range from about 10 to 30cm
with the system used by the HIRELADO group. A
shorter distance may saturate the detector unit and
a longer distance results in impaired signal-to-noise
mendeddistance issomewhat dependenton thecol-
our of the tissue, but a target distance of about 15–
17cm will generally work well for most tissues. If
there is a tendency for detector saturation, this will
show up as uniform white areas in the photo image,
while too low a detected light level will generally re-
sult in areas clearly belonging to the object being
classified as background.
To compare the perfusion values within areas of
interest in two images it is important to use the
same distance for the image capturing procedures.
This can be achieved by use of a laser pen fixed to
the scanner head, setting its beam to cross the
LDPI laser beam at about a 45æ angle at the target
distance. In consecutive measurements the same
distance is easily maintained by letting the two
laser beams coincide to form a single spot on the
tissue surface. The additional laser beam must be
switched off during the image capturing procedure.
The optimal working distance can be assessed by
recording test images at different distances.
If a single image of a tissue area with essentially
stationary perfusion is to be captured, the Image
mode may be set to Single. When a slow change in
perfusion over time is expected, with accompanying
involvement of the extent of the area over which the
Repeated. This setting enables the automatic re-
cording of many images over an extended period of
time. If more rapid changes in tissue perfusion fol-
tial mode displaying a number of small subimages
within the same image window may be the best
choice. When a time trace recording of tissue per-
fusion in a single or small number of measurement
sites is of primary interest, the best choice of setting
tem allows for averaging the recorded tissue per-
fusion within an area, the extent of which is set by
the user, thereby eliminating the main shortcoming
of the single-point laser Doppler flowmeter.
Image size and resolution
The size of the image should be set by use of the
Image size control (determining the number of
measurement sites in the image) or by the Resol-
ution control (determining the zoom-in and zoom-
out effect making the overall tissue area covered
smaller or larger at a given number of total meas-
urement sites), rather than by changing the dis-
tance between scanner head and tissue. A full for-
mat image is composed of 64¿64 perfusion values
and is recorded in a matter of minutes, the exact
time dependent on the Measurement speed (see be-
low). The optimal choice of Image size and Resol-
ution depends on the application. The resolution
limit is further determined by the laser-beam diam-
eter and the distance between two consecutive
measurement sites. With a laser-beam diameter of
0.6mm the smallest objects that can be resolved
are of the dimension 1.2mm. LDPI technologies
that employ focused laser beams in order to im-
prove further the resolution, and systems that can
generate larger images for flow mapping of, for ex-
ample, the entire upper part of the body are avail-
able on the market.
In the system used by the HIRELADO group, the
beam is kept still in relation to the tissue for a cer-
136FULLERTON ET AL.
tain time period (about 50ms in Standard Speed
setting, compatible with a lower frequency cut-off
frequency of 20Hz) at each measurement site.
With this setting about 4min is required to capture
a full format image including altogether 4096 per-
fusion values. To be able to obtain overview images
at higher speed, the capturing time can be reduced
to about 90s, albeit at the expense of image quality
(higher noise levels). It is recommended that the
Standard Speed setting be used if image integrity
is of primary importance. The overall image cap-
turing time can best be kept to a minimum by set-
ting the Image size to the smallest possible area
covering the tissue region of interest.
The Background threshold determines at what total
light intensity the actual measurement point
should be classified as belonging to the back-
ground. In order to maintain the best background
discrimination, it is therefore recommended that
the object be placed on a light absorbing back-
ground, such as a green cloth. By adjusting the
Background threshold level to comply with the
tissue colour, distance to tissue and ambient light
level, adequate background discrimination is gen-
erally possible to achieve in most situations.
Colour scale setting
During the capturing procedure, the Colour scale
should preferably be set to Relative colour scale,
because this will calculate an individual colour
scale, stretched between the lowest and highest per-
fusion value, in each image. The Colour scale can
easily be altered in the image analysis process and
made identical for a given set of images, whereas
the recorded perfusion values that form the basis
of average value calculations within selected areas
of interest cannot be altered by the user.
The Mark area control makes the laser beam move
quickly along the boundaries of the image to be
captured. This feature is most useful in determin-
ing the appropriate size of the image and in posi-
tioning the scanner head over the tissue.
Capturing an image
During image capturing it is important that the
test subjects/patients, after having been informed
that no movement is allowed during the image cap-
turing procedure, rest in a comfortable position in
order not to influence the tissue blood perfusion to
be investigated. It is furthermore of fundamental
importance that all conversation with the test sub-
ject/patient is abandoned during the image captur-
ing process. The skin must not be touched immedi-
ately prior to or during image capturing, since the
slightest mechanical stimuli may affect the micro-
circulation. Slow movements, such as thoracal
breathing movements, do normally not interfere
with the Doppler signal. It is strongly recom-
mended that the image is stored immediately after
completion of the image capturing process.
If the result of a provocation is to be quantita-
tively assessed, it must be borne in mind that the
numerical values of the perfusion calculated may
differ if the area of interest is in the centre or at
the periphery of the image, and if the skin surface
is perpendicular to the beam whether it is in its
central position or not. Recording test images with
the area of interest in the centre and at the peri-
phery and comparing the average perfusion values
of these areas can easily test the influence of these
Evaluation and interpretation
In this section, some generally applicable basic as-
pects of image evaluation and interpretation are
addressed. Evaluation and interpretation in some
specific applications of particular importance are
dealt with separately in the next section.
When different images are to be compared they
must be displayed in the same colour scale. This
is achieved by first identifying the highest single
perfusion value in any of the set of images to be
compared or any other arbitrary value of choice,
and then using this value as the maximum per-
fusion in the User defined colour scale for each of
the images. The minimum value can in a similar
way be set to the lowest perfusion value in any of
the images or, alternatively, to zero.
By selecting a Region of Interest (ROI) in a per-
fusion image, the statistics (including mean value,
standard deviation and number of measurement
sites) of the ROI can be calculated. This infor-
mation forms the basis for comparing the per-
fusion of two areas (in the same or different im-
ages) from a statistical point of view. The large
number of measurement sites generally makes it
possible to demonstrate statistically (at a reason-
able level of significance) a difference in average
perfusion, given that this difference exists in reality.
The images displaying the detailed spatial vari-
ability in tissue perfusion can thus conveniently be
transferred to numbers and figures of statistical
significance, in contrast to the situation with
single-point laser Doppler monitoring.
There are numerous pitfalls in interpreting areas
137 LASER DOPPLER IMAGING
with deviating perfusion values within an image,
many of which can easily be revealed and iden-
If the test subject/patient moves during the im-
age capturing procedure, a distinct horizontal strip
generally shows up in the perfusion image. This
movement artefact is caused by general movement
of the tissue, which the LDPI system cannot easily
separate from the specific movements of blood
cells. Movement artefacts can easily be recognized,
however, since they generally occupy only a single
(or a multitude of) horizontal line(s) in the image.
If the tissue surface is covered with a liquid film,
face, generally resulting in an isolated spot of alter-
ed (in most cases reduced) perfusion values. This
artefact can generally also be readily identified,
since the corresponding photo image is plagued
with a corresponding white spot, indicating locally
intense light reflection in the liquid film.
Since the perfusion image is captured over a time
period sometimes of several minutes, it must be
assumed that the tissue perfusion during this time
interval is stationary. If not, the temporal vari-
ability in tissue perfusion appearing on a shorter
time scale may show up falsely as spatial differ-
ences in perfusion. These artefacts are generally
more difficult to identify, but regular patterns in
perfusion with intervals corresponding to respir-
atory cycles, vasomotion and even the pulse may
be generated by temporal rather than spatial vari-
ations in tissue perfusion.
Summary of practical guidelines
Prior to image capturing
O See ‘Guidelines for measurement of cutaneous
blood flow by laser Doppler flowmetry’ for an
overview on how to prepare a laser Doppler-
based recording session.
O Pay special attention to:
Subject should sit/lie in a comfortable position.
Acclimatization for 20–30min with test sites left
O Keep ambient light level low.
O Avoid ‘noisy’ environments.
O Ambient temperature influences skin blood per-
O Carefully study the instrument manual for dif-
ferent set-up alternatives.
O Perform the calibration check on a daily basis.
O Select the optimal scanner headªtissue dis-
O Mark the area with ink spots or tape.
O Handle the skin carefullyªavoid pressure on
and scratches of the skin area under study.
O Stationary blood perfusion: if applicable, in-
clude a normal skin reference area in the image
O Non-stationary blood perfusion: record a series
of small images to demonstrate temporal per-
During image capturing
O Avoid unnecessary communication with the test
O Avoid unnecessary activities in the examination
O Do not touch the instrument (or the tissue) dur-
ing the recording procedure.
O Watch out for unintentional subject/patient
After image capturing
O Insert comments in the Note Pad accompanying
O Inspect images for movement artefacts (hori-
O Save the image.
O Select the Region of Interest and calculate aver-
O To compare different images colour-wise they
need to be set to the same colour scale.
O Save the postprocessed images.
O If applicable, convert images to TIFF-file for-
mat and combine them in, for example, a
Word document for overview presentation and
Present use and future potential
LDPI technology has so far been used in derma-
tology, diabetology, pharmacology, wound healing
and numerous other applications. Of particular in-
terest for the research groups involved in the HIR-
ELADO project are the use of LDPI in skin al-
lergy and irritant patch testing, microdialysis and
skin tumour blood perfusion evaluation. These ap-
plications are described in further detail below.
Skin allergy and irritant patch testing
A major dermatological application of LDPI is in
the evaluation of irritant and allergic skin reac-
tions, in which a high sensitivity and reproducibil-
138FULLERTON ET AL.
Table1. Papers reporting skin irritant and allergy patch testing
of various substances
allergy (Kathon CG) (18)
allergy (nickel sulphate) (19)
allergy (corticosteroid) (20)
allergy (budesonide) (21)
principle (17, 22, 23)
allergy (diesel oil) (24)
allergy (sodium dodecyl sulphate) (25)
irritant (D vitamins) (26)
irritation (calcipotriol) (27)
irritation (methyl nicotinate) (28)
allergy (gold) (29)
atopic dermatitis (30)
image analysis (31)
ity have been demonstrated (see Table1). Such in-
vestigations are best performed by use of an LDPI
system with relatively high resolution. In such im-
ages the increased blood perfusion is caused by the
hyperperfusion in inflamed and vasodilated skin.
LDPI does not, however, directly distinguish be-
tween allergic and irritant reactions.
The gold standard for evaluation of skin tests is
visual scoring. LDPI evaluation of epicutaneous
and intracutaneous tests constitutes an objective
and practically useful method for the quantifi-
cation of such test reactions. LDPI values of visu-
ally positive tests allow a calculation of a lower
limit of positive LDPI tests. Visually negative tests
usually have low LDPI values that differ signifi-
cantly from those obtained from positive readings.
There is generally a high correlation between visual
scoring and LDPI evaluation. The LDPI values
tend to increase slightly before the visual reactions
can be noticed and the high LDPI values last for
a somewhat shorter time period than the positive
visual reactions. Thus, reactions associated with
high LDPI values, which are visually negative, as
a rule later become visually positive. Visually posi-
tive test reactions with LDPI values below the cal-
culated positive limit are as a rule late test readings
of previously visually and LDPI-positive reactions.
The interesting ‘grey zone’ of questionable reac-
tions with LDPI readings between the negative and
the positive LDPI limit needs to be further ana-
Repeated and comparative reading of tests
necessitates the use of control areas to overcome
long- term temporal variations in vascular reac-
tions and the influence of temperature. The use of
two control patches with vehicle alone, constitutes
an efficient way to increase the accuracy of the test
The optimal size of a skin test image is some-
what dependent on the application and on the sub-
stance to be tested. Normally, the size of the skin
area investigated should equal the size of the
patch, plus 1–2cm around its edges. In most cases
this coincides with a square with the side equal to
three times the diameter of the patch. With very
strong reactions, an increased blood perfusion may
be found in an area of six times the size of the
patch. If possible, the area of the central patch size
should include at least 100 measurement sites. This
complies with a standard Finn Chamber of 9mm
diameter and a laser spot size of about 1mm2.
If the test patches are placed on the back, the
test subject/patient should lie comfortably in the
prone position. When the patches are placed on
the forearm, the test subject/patient should sit
comfortably in a chair with the test area at heart
level. In the latter case, the patches should not be
placed directly over any major vein, because the
blood flow in those vessels will contribute to the
recorded Doppler signal.
One method of evaluating a test is to find the
mean increase in perfusion of the central area of
the test reaction equal to the size of the patch. For
intracutaneous tests, an area size of 1cm2is satis-
factory. A more exact but also more time-consum-
ing method is to evaluate the increase in blood per-
fusion of the test area through a ‘volume inte-
gration’ of a larger area. With this method the
total area of increased blood perfusion has to be
compared with the control area(s). For this pur-
pose it is necessary to evaluate an area of about 3
¿3cm. An extremely strong reaction may, how-
ever, cover an area of 6¿6cm or more.
Detailed recommendations, based on more than
25,000 allergic patch test assessments, on how to
standardize the conditions for this particular ap-
plication were compiled by Bjarnasson et al. in
Table1 includes some of the early publications
in the field of allergy and irritant patch testing
using the LDPI technology.
Microvascular aspects of skin microdialysis are
best studied by use of a very high resolution LDPI
system (based on a focused laser beam to improve
the resolution) that operates at a distance of 7cm,
giving a total field view of only 1.5¿1.5cm (32–
34). This device is well suited for localization of
the microdialysis probe in such perfusion experi-
ments (35, 36) and helps to interpret the results
obtained with this probe. In addition, it is possible
to investigate the settling time of blood perfusion
caused by the insertion of the probe and, when
applicable, to use this information to trigger the
sampling of substances by use of the microdialysis
139 LASER DOPPLER IMAGING
probe, in order to avoid influences of increased
skin blood perfusion on the results.
Skin tumour blood perfusion
Studies on skin tumours and other skin lesions
constitute promising new applications for LDPI
investigations. Port wine stains have been evalu-
ated before and after laser treatment (37). The tu-
mour circulation in association with photodynam-
ic therapy of non-melanoma skin tumours has
been followed over the treatment period (38). A
recently published method to distinguish moles
from melanomas with a high-resolution scanner
(39) constitutes a promising approach for future
applications in the differential diagnosis of skin tu-
mours. In addition to simply recording the average
perfusion within the tumour boundary or exten-
sion of the hyper-perfused tumour area, more ad-
vanced image processing methods may be em-
ployed to further refine the opportunity for tu-
mour differential diagnosis offered by LDPI
technology. These image-processing methods in-
clude feature extraction, analysis of boundary ir-
regularities and measures of intratumour blood
In addition to the applications above, which have
been addressed within the framework of the HIR-
ELADO project, LDPI technology has been suc-
cessfully applied in a number of other fields includ-
ing: diabetes neuropathy (cold provocation) (40–
42), endothelial cell malfunction in diabetes (ion-
tophoresis of acetylcholine) (43, 44), leg ulcers and
wound healing(drug influences)
models) (48, 49), psoriasis (50, 51), scleroderma
and Raynaud’s disease (52, 53)and flap surveil-
lance (54, 55).
The present work was supported by the European Com-
mission DG XIII/DªInnovation programme (contract
1. Stern M D. In vivo evaluation of microcirculation by coher-
ent light scattering. Nature 1975: 254: 56–58.
2. Holloway G A Jr, Watkins D W. Laser Doppler measure-
ment of cutaneous blood flow. J Invest Dermatol 1977: 69:
3. Nilsson G E, Tenland T, Öberg P Å. A new instrument
for continuous measurement of tissue blood flow by light
beating spectroscopy. IEEE Trans Biomed Eng 1980: 27:
4. Leahy M J, de Mul F F, Nilsson G E, Maniewski R. Prin-
ciples and practice of the laser-Doppler perfusion tech-
nique. Technol Health Care 1999: 7: 143–162.
5. Line P D, Mowinckel P, Lien B, Kvernebo K. Repeated
measurement variation and precision of laser Doppler
flowmetry measurements. Microvasc Res 1992: 43: 285–293.
6. Nilsson G E, Jakobsson A, Wårdell K. Imaging of tissue
blood flow by coherent light scattering. In: 11th Ann EMBS
conference. Seattle: IEEE, 1989.
7. Wårdell K, Jakobsson A, Nilsson G E. Laser Doppler per-
fusion imaging by dynamic light scattering. IEEE Trans
Biomed Eng 1993: 40: 309–316.
8. Essex T J, Byrne P O. A laser Doppler scanner for imaging
blood flow in skin. J Biomed Eng 1991: 13: 189–194.
9. Fujii H, Nohira K, Yamamoto Y, Ikawa H, Ohura T.
Evaluation of blood flow by laser speckle image sensing.
Applied Optics 1987: 26: 5321–5325.
10. Bircher A, de Boer E M, Agner T, Wahlberg J E, Serup J.
Guidelines for measurement of cutaneous blood flow by
laser Doppler flowmetry. A report from the Standardiza-
tion Group of the European Society of Contact Dermatitis.
Contact Dermatitis 1994: 30: 65–72.
11. Fullerton A, Fischer T, Lahti A, Wilhelm K P, Takiwaki H,
Serup J. Guidelines for measurement of skin colour and
erythema. A report from the Standardization Group of the
European Society of Contact Dermatitis. Contact Derma-
titis 1996: 35: 1–10.
12. Groner W, Winkelman J W, Harris A G et al. Orthogonal
polarization spectral imaging: a new method for study of
the microcirculation. Nature Med 1999: 5: 1209–1212.
13. Bollinger A, Butti P, Barras J P, Trachsler H, Siegenthaler
W. Red blood cell velocity in nailfold capillaries of man
measured by a television microscopy technique. Microvasc
Res 7 1974: 61.
14. Harrison D K, Abbot N C, Beck J S, McCollum P T. Laser
Doppler perfusion imaging compared with light-guide laser
Doppler flowmetry, dynamic thermographic imaging and
tissue spectrophotometry for investigating blood flow in
human skin. Adv Exp Med Biol 1994: 345: 853–859.
15. Andersson R R, Parrish J A. The optics of human skin. J
Invest Dermatol 1981: 77: 13–19.
16. Jakobsson A, Nilsson G E. Prediction of sampling depth
and photon pathlength in laser Doppler flowmetry. Med
Biol Engineering Computing 1993: 31: 301–307.
17. Bjarnason B, Flosadottir E, Fischer T. Objective non-invas-
ive assessment of patch tests with the laser Doppler per-
fusion scanning technique. Contact Dermatitis 1999: 40:
18. Bjarnason B, Fischer T. Visual and laser Doppler perfusion
scanning assessments of Kathon CG patch test reactions.
Am J Contact Dermatitis 1998: 9: 224–230.
19. Bjarnason B, Fischer T. Objective assessment of nickel sul-
fate patch test reactions with laser Doppler perfusion im-
aging. Contact Dermatitis 1998: 39: 112–118.
20. Bjarnason B, Flosadottir E, Fischer T. Reactivity at edges
of corticosteroid patch tests may be an indicator of a strong
positive test response. Dermatology. 1999: 199: 130–134.
21. Bjarnason B, Flosadottir E, Fischer T. Assessment of
budesonide patch tests. Contact Dermatitis 1999: 41: 211–
22. Fullerton A, Serup J. Laser Doppler image scanning for
assessment of skin irritation. In: Elsner P, Maibach H (eds.)
Irritant Dermatitis. New Clinical and Experimental Aspects.
Current Problems in Dermatology. Basel: Karger, 1995:
23. Stücker M, Auer T, Hoffman K, Altmeyer P. Two-dimen-
sional blood flow determinations in allergic reactions using
laser Doppler scanning. Contact Dermatitis 1995: 33: 299–
140 FULLERTON ET AL. Download full-text
24. Fischer T, Bjarnason B. Sensitizing and irritant properties
of 3 environmental classes of diesel oil and their indicator
dyes. Contact Dermatitis 1996: 34: 309–315.
25. Fischer T, Dahlen A, Bjarnason B. Influence of patch-test
application tape on reactions to sodium dodecyl sulfate.
Contact Dermatitis 1999: 40: 32–37.
26. Fullerton A, Serup J. Topical D-vitamins. multiparametric
comparison of the irritant potential of calcipotriol, tacalci-
tol and calcitriol in a hairless guinea pig model. Contact
Dermatitis 1997: 36: 184–190.
27. Fullerton A, Benfeldt E, Petersen J R, Jensen S B, Serup J.
The calcipotriol dose–irritation relationship: 48 hour oc-
clusive testing in healthy volunteers using Finn Chambers.
Br J Dermatol 1998: 138: 259–265.
28. Issachar N, Gall Y, Borrel M T, Poelman M C. Correlation
between percutaneous penetration of methyl nicotinate and
sensitive skin, using laser Doppler imaging. Contact Der-
matitis 1998: 39: 182–186.
29. Moller H, Bjorkner B, Bruze M, Lundqvist K, Wollmer P.
Laser Doppler perfusion imaging for the documentation of
flare-up in contact allergy to gold. Contact Dermatitis 1999:
30. Stucker M, Heese A, Hoffmann K, el Gammal S, Altmeyer
P. Quantification of vascular dysregulation in atopic derma-
titis using laser Doppler perfusion imaging. Skin Res Tech-
nol 1998: 4: 9–13.
31. Wårdell K, Andersson T, Anderson C. Analysis of laser
Doppler perfusion images of experimental irritant skin re-
actions. Skin Res Technol 1996: 2: 149–157.
32. Anderson C, Andersson T, Wårdell K. Changes in skin cir-
culation after insertion of a microdialysis probe visualized
by laser Doppler perfusion imaging. J Invest Dermatol
1994: 102: 807–811.
33. Andersson T, Wårdell K, Anderson C. Human in vivo cu-
taneous microdialysis: estimation of histamine release in
cold urticaria. Acta Derm-Venereol 1995: 75: 343–347.
34. Krogstad A L, Jansson P-A, Gissle ´n P, Lönnroth P. Micro-
dialysis methodology for assessment of dermal fluid in
humans. Br J Dermatol 1996: 134: 1005–1012.
35. Linde ´n M, Wårdell K, Andersson T, Anderson C. High
resolution laser Doppler perfusion imaging for the investi-
gation of blood circulatory changes after microdialysis
probe insertion. Skin Res Technol 1997: 3: 227–232.
36. Groth L, Serup J. Cutaneous microdialysis in man. effects
of needle insertion trauma and anaesthesia on skin per-
fusion, erythema and skin thickness. Acta Derm-Venereol
1998: 78: 5–9.
37. Troilius A M, Ljunggren B. Evaluation of port wine stains
by laser Doppler perfusion imaging and reflectance pho-
tometry before and after pulsed dye laser treatment. Acta
Derm-Venereol 1996: 76: 291–294.
38. Wang I, Andersson-Engels S, Nilsson G E, Wårdell K,
Svanberg K. Superficial blood flow following photodynam-
ic therapy of malignant non-melanoma skin tumours meas-
ured by laser Doppler perfusion imaging. Br J Dermatol
1997: 136: 184–189.
39. Stücker M, Horstmann I, Nuchel C, Rochling A,
Hoffmann K, Altmeyer P. Blood flow compared in benign
melanocytic naevi, malignant melanomas and basal cell
carcinomas. Clin Exp Dermatol 1999: 24: 107–111.
40. Arora S, Smakowski P, Frykberg R G, Simeone L R, Free-
man R, LoGerfo F W, Veves A. Differences in foot and
forearm skin microcirculation in diabetic patients with and
without neuropathy. Diabetes Care 1998: 21:1339–1344.
41. Bornmyr S, Castenfors J, Svensson H, Wroblewski M,
Sundkvist G, Wollmer P. Detection of autonomic sym-
pathetic dysfunction in diabetic patients. A study using
laser Doppler imaging. Diabetes Care 1999: 22: 593–597.
42. Bornmyr S, Svensson H, Lilja B, Sundkvist G. Cutaneous
vasomotor responses in young type I diabetic patients. J
Diabetes Complications 1997: 11: 21–26.
43. Morris S J, Shore A C, Tooke J E. Responses of the skin
microcirculation to acetylcholine and sodium nitroprusside
44. Veves A C M, Akbari J, Primavera V M et al. Endothelial
dysfunction and the expression of endothelial nitric oxide
synthetase in diabetic neuropathy, vascular disease, and
foot ulceration. Diabetes 1998: 47: 457–463.
45. Gschwandtner M E, Ambro ´zy E, Böhler K et al. Reprodu-
cibility of laser Doppler imager flux measurements within
ischemic or venous ulcers and adjacent skin. Skin Res Tech-
nol 1998: 4: 222–227.
46. Gschwandtner M E, Ambrozy E, Schneider B, Fasching S,
Willfort A, Ehringer H. Laser Doppler imaging and capil-
lary microscopy in ischemic ulcers. Atherosclerosis 1999:
47. Mayrovitz H N, Smith J. Heel-skin microvascular blood
perfusion responses to sustained pressure loading and un-
loading. Microcirculation 1998: 5: 227–233.
48. Couffinhal T, Silver M, Zheng L P, Kearney M, Witzen-
bichler B, Isner J M. Mouse model of angiogenesis. Am J
Pathol 1998: 152: 1667–1679.
49. Rivard A, Fabre J E, Silver M et al. Age-dependent impair-
ment of angiogenesis. Circulation 1999: 99: 111–120.
50. Auer T, Bacharach-Buhles M, el Gammal S et al. The
hyperperfusion of the psoriatic plaque correlates histolog-
ically with dilatation of vessels. Acta Derm Venereol
(Stockh) 1994: 186 (Suppl): 30–32.
51. Krogstad A L, Lonnroth P, Larson G, Wallin B G. Capsaic-
in treatment induces histamine release and perfusion
changes in psoriatic skin. Br J Dermatol 1999: 141: 87–93.
52. Aghassi D, Monoson T, Braverman I. Reproducible meas-
urements to quantify cutaneous involvement in sclero-
derma. Arch Dermatol 1995: 131: 1160–1166.
53. Stücker M, Jeske M, Hoffmann K, Altmeyer P. Laser-
Doppler-Scanning bei progressiver systemischer Skleroder-
mi. Phlebologie 1995: 24: 9–14.
54. Arnold F, He C F, Jia C Y, Cherry G W. Perfusion imaging
of skin island flap blood flow by a scanning laser-Doppler
technique. Br J Plastic Surg 1995: 48: 280–287.
55. Eichhorn W, Auer T, Voy E D, Hoffmann K. Laser Doppler
imaging of axial and random pattern flaps in the maxillo-
facial area. A preliminary report. J Cranio-Maxillo-Facial
Surg 1994: 22: 301–306.