Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
A simultaneous multimodal imaging system for tissue functional
parameters
Wenqi Ren1, Zhiwu Zhang1, Qiang Wu1, Shiwu Zhang1*, Ronald Xu1, 2 1Centers for Biomedical Engineering, University of Science and Technology of China, Hefei,
230027, China; 2Department of Biomedical Engineering, the Ohio State University, Columbus, OH
43210
ABSTRACT
Simultaneous and quantitative assessment of skin functional characteristics in different modalities will facilitate diagnosis
and therapy in many clinical applications such as wound healing. However, many existing clinical practices and
multimodal imaging systems are subjective, qualitative, sequential for multimodal data collection, and need co-registration
between different modalities. To overcome these limitations, we developed a multimodal imaging system for quantitative,
non-invasive, and simultaneous imaging of cutaneous tissue oxygenation and blood perfusion parameters. The imaging
system integrated multispectral and laser speckle imaging technologies into one experimental setup. A Labview interface
was developed for equipment control, synchronization, and image acquisition. Advanced algorithms based on a wide gap
second derivative reflectometry and laser speckle contrast analysis (LASCA) were developed for accurate reconstruction
of tissue oxygenation and blood perfusion respectively. Quantitative calibration experiments and a new style of skin-
simulating phantom were designed to verify the accuracy and reliability of the imaging system. The experimental results
were compared with a Moor tissue oxygenation and perfusion monitor. For In vivo testing, a post-occlusion reactive
hyperemia (PORH) procedure in human subject and an ongoing wound healing monitoring experiment using dorsal
skinfold chamber models were conducted to validate the usability of our system for dynamic detection of oxygenation and
perfusion parameters. In this study, we have not only setup an advanced multimodal imaging system for cutaneous tissue
oxygenation and perfusion parameters but also elucidated its potential for wound healing assessment in clinical practice.
Keywords: Simultaneous, multispectral, laser speckle, oxygenation, perfusion, phantom, nude mice, wound healing.
* Corresponding author: [email protected], tel: 86-0551- 63601482
1. INTRODUCTION
Chronic wounds are among the most common wounds in clinical medicine, which greatly reduce patients’ quality of life
and cost hundreds of billion on the treatment every year [1]. A variety of etiologic, systemic, and local factors may be
involved in the pathogenesis of a chronic wound. Normal wound healing process involves the reparative phases of
inflammation, proliferation, and remodeling [2]. Interrupting any of these phases may result in chronically unhealed
wounds, amputation, or even death. Therefore, accurate and quantitative characterization of the functional parameters at
each phase of the wound healing process will help objectively assess the clinical outcome and quantitatively guide the
therapeutic process.
Multimodal Biomedical Imaging IX, edited by Fred S. Azar, Xavier Intes, Proc. of SPIE Vol. 8937, 893706 · © 2014 SPIE · CCC code: 1605-7422/14/$18 · doi: 10.1117/12.2041008
Proc. of SPIE Vol. 8937 893706-1
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
ALaser light source
Electronic shutter
CCDcamera
BShutter Speed(ros,
1,1250 SOO
A ¡0 SCO 1200
wavelengtl, pre.egro pre On
AOTF light source"-Opt i ca I fiber
Light ring
Intensity ..01*%NM
255.00
Npe,wsh, 5w-1
Dark Images Pat, AHyperSpectral Images Path
E01109,,),
cumem:nte,a, gm, wave,. 1,009
0
..sse.Sgetkie Images Pad,
laj
5, 7
She., Speeg
"4"
da, make
CZ* C:0
t
____!) 7
ShuaerS
peed(ms)
0250
S00
Gain
8000
5001200
wavelength pre -expo
pre -gain
susM
ka Iinterval(m
sl surreniexposing
15000
Intensity255.00
Doom
Limit
PE
CT
OT
Up lim
itP
roper intensity,r
1I
6P9
Auto-E
xpo(E
nd Expo I
HyperSpectralineges C
atS
F:t11001T
;F
ca,rent..aw
4ngthh pp,ng
0
,rsrval(msl
:; 1200
.aserSpecIde Im
ages path
SE
\1109\\Iib
intervagms)
4ser;500
Geie
Shutter Speed200
10
Tim
<art,
Oxygen exists in biological tissue in multiple forms, accurate assessment of wound tissue oxygenation is important for
appropriate diagnosis, prevention, and treatment of chronic wounds [3]. Recently, multispectral imaging opens a new
avenue for non-invasive and real-time assessment of tissue oxygenation. Images of the tissue at multiple wavelengths were
collected and a number of methods have been investigated to reconstruct the oxygen saturation. Besides oxygenation,
sufficient nutritive delivered by subcutaneous perfusion is also critical for wound healing process, and the blood vessel
evolution is also involved in all the three healing phases [4]. At present, as a full-field technique that produces two-
dimensional map of blood flow over an area of tissue, Laser Speckle Contrast Analysis (LASCA) based on the spatial or
temporal statistic of speckle pattern has been commonly used for clinical assessment.
The development of all imaging methods requires the use of tissue-simulating phantoms to mimic the properties of tissues.
Generally, phantoms are used for testing system designs, optimizing signal to noise in existing systems, performing routine
quality control, and comparing performance between systems [5]. In this study, a liquid blood phantom consists of fresh
blood, India ink, and intralipid was prepared for multispectral imaging of oxygen saturation. An integrated calibration
method using a Flux Standard and a moving white diffuser was developed for accurate laser speckle imaging of blood
perfusion. In addition, we designed a new kind of reusable phantom made of polydimethylsiloxane (PDMS), which could
be used to simulate the optical characteristics and blood circulation in cutaneous tissue.
Despite the progress in tissue oxygenation and perfusion measurements, no imaging tool is available yet for simultaneous
assessment of these two important parameters in a uniform field of view (FOV).In this paper, we developed a multimodal
imaging system that integrated multispectral imaging and laser speckle imaging modules, multimodal images were
acquired synchronized with corresponding light illumination switched by an electronic shutter. Tissue oxygenation and
blood perfusion were reconstructed by a wide gap second derivative reflectometry and laser speckle contrast analysis
(LASCA) algorithm respectively. Quantitative calibration experiments and skin-simulating phantoms were designed to
verify the numerical accuracy and reliability of the imaging system. In vivo, a post-occlusion reactive hyperemia (PORH)
procedure, and an ongoing wound healing process monitoring experiment using nude mice dorsal window chamber models
were conducted to demonstrate the potential of our system for wound healing assessment in clinical practice.
2. MULTIMODAL IMAGING SYSTEM
Figure 1. (A) Integrated multimodal imaging system. (B) User interface of the imaging system
A B
Proc. of SPIE Vol. 8937 893706-2
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
The multimodal imaging system integrates multispectral and laser speckle imaging technologies into one setup. The
greatest advantage of this system is the combination of the illumination and images acquisition from two different imaging
modalities, which can realize the identical field of view and simultaneous data acquisition. The components of the system
are shown in Figure 1 (A), an AOTF tunable light source (500~900nm range, 10nm bandwidth@633nm, Brimrose, US)
and a laser device (λ=785nm, 150mW, Changchun New Industries, China) are connected to a light ring via a bifurcated
optical fiber for average illumination, an electronic shutter (Daheng New Epoch Technology, Inc. Beijing, China) is fixed
on the fiber to switch the two light source periodically. The reflected intensity of tissue or phantoms will be imaged onto
a homochromous 12-bit CCD camera (Microvision Digital Imaging Technology Co.Ltd, China) with 1392×1040 pixels.
The quantum efficiency of the CCD chip is greater than 50% at a wavelength range from 400~650nm, suitable for
multispectral imaging. The light source and electronic shutter are connected to a laptop via serial port, and the camera is
connected through 1394A port. To control all the equipment and synchronize the data acquisition with modalities switching,
we developed a Labview (National Instruments, TX, US) interface as shown in figure 1(B). The program has a function
of automatic exposure in case of improper image intensity. With the integrated multimodal imaging system, the
noninvasive, simultaneous, and multimodal measurement of the tissue characteristics can be realized.
3. ALGORITHMS AND METHODS OF THE IMAGING SYSTEM
3.1. Multispectral imaging of tissue oxygenation
Reflection spectrum of a semi-infinite turbid media was simulated using a simplified numerical model [6]. In our system,
tissue oxygen saturation (StO2) was reconstructed by a wide gap second derivative reflectometry [7]. Six wavelengths (544,
552, 568, 576, 592, 600nm) were selected to calculate a Second Derivative Ratio (SDR), which is an index only corrected
to StO2 regardless of blood concentration, scattering and melanin level. SDR was calculated as:
SDR =)576(2)552()600(
)568(2)544()592(
rrr
rrr
, r (λ) =
)(
)(
s'
a
(1)
where μa is the absorption coefficient and μs is the reduced scattering coefficient. The index has a monotonous relationship
with StO2, thus one can calculate StO2 from the measurements of SDR.
The procedure of multispectral imaging for oxygenation StO2 using various wavelengths is introduced below. Firstly,
sample to be measured and a 99% reflected diffuser (The National Institute of Standards and Technology, Gaithersburg,
MD) were placed within the FOV of the CCD camera at an equal height. Then an image in totally dark environment was
captured to record the dark current noise of the CCD chip. Next, the AOTF tunable light source was controlled to illuminate
the specific wavelength beam successively, meanwhile the auto-exposure function was run to set up the optimal exposure
time t of each wavelength in case of improper intensity. After that, the monochromatic images of every wavelength will
be taken and stored at corresponding exposure time for further processing by the wide gap second derivative reflectometry
as described above. In addition, to avoid the measurement artifact caused by ambient light, it was suggested that the
experiment should be conducted in a dark environment.
3.2. Laser speckle imaging of tissue perfusion
Tissue perfusion characteristics can be acquired with Laser Speckle Contrast Analysis (LASCA) [8]. Illumination of a
tissue by monochromatic laser light will produce an interference pattern on the tissue surface. When the illuminated object
Proc. of SPIE Vol. 8937 893706-3
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
is static, the speckle pattern is stationary. However, when moving particles (such as blood cells) are present in the sampled
tissue, the speckle pattern will fluctuate over time. By analyzing these intensity fluctuations, parameter about the blood
perfusion in the tissue can be obtained.
In this system, a 785nm laser device was selected for enough depth of light transmission, images of tissue speckle pattern
were recorded by camera synchronizing with the electronic shutter switch to laser illumination. From the images data, we
can get the intensity and variance of a region of interest (ROI). To calculate the contrast from intensity – variance pairs,
the following formula should be used:
C=I
V (2)
where C is the contrast, V is the variance, I is the intensity and β is the coherence factor. The coherence factor ensures C =
1 (and thus perfusion = 0) for static objects. It is system specific and will be determined by a calibrate experiment as
described later. The perfusion is calculated from the contrast as follows:
P= )11( Ck (3)
where P is the perfusion, C the contrast and k the signal gain factor. The same as the coherence factor, the signal gain
factor will be calibrated and ensure a certain perfusion values on a flux standard.
4. PHANTOM EXPERIMENTS AND SYSTEM CALIBRATION
4.1. Blood Phantom experiment for multispectral imaging
4.1.1 Blood Phantom Fabrication and calibration methods
The StO2measurement was verified using a liquid blood phantom [9], which was prepared by mixing fresh chicken blood,
India ink (Bomei Biotechnology, Hefei, China), and 20% intralipid (Sino-Swed Pharmaceutical Corp, Ltd. China) in
phosphate buffered solution (Bomei Biotechnology, Hefei, China).Ink and intralipid were used to simulate the melanin
absorption and skin scattering respectively, and the concentration of these ingredients should be determined to ensure the
optical parameters of the phantom would match with those of in vivo tissue. In our experiments, the absorption coefficients
of the blood phantom was regulated to be 0.172 cm-1 (at 690 nm) and the reduced scattering coefficient to be 4.5 cm-1 (at
690 nm), as confirmed by an OxiplexTS tissue spectrophotometer (ISS Inc., Champaign, IL). The pH level of the phantom
was adjusted to 7.4 by adding NaOH and HCl. Finally, The oxygenation level of the phantom was adjusted by dropwise
addition of sodium hydrosulfite (0.025 g/mL, Bomei Biotechnology, Hefei, China) using an injection syringe (LSP01-1A,
Baoding Longer Precision Pump Co., Ltd, China).
Proc. of SPIE Vol. 8937 893706-4
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
CCDcamera AOTF light source
Syringe pump
c )
Moor oxygenationmonitor
liii1.1
NIST traceablewhite diffuser
cao,ÇoE)Lco
100
Erói 80oo
úaa 60fA
7ET
40aUlWC
E 20NO«N
o,
O Normalized St02
linear fitting
20 40 60 80 100St02( %)measured by Moor Oxygenation monitor
0:
Figure 2. Schematic drawing of an experimental setup for multispectral imaging of a blood phantom.
To calibrate the multispectral imaging method, multispectral data of the blood phantom were acquired from a setup as
shown in Figure 2. The blood phantom was placed in a container full of argon gas to eliminate the measurement artifact
caused by oxygen in ambient air. During the calibration test, 11oxygenation levels ranging from nearly 100% to nearly 0%
were achieved by dropwise addition of sodium hydrosulfite at proper doses. At each oxygenation plateau, the blood
phantom was mixed to homogeneity by a magnetic stirrer, then multispectral data of each oxygenation level were acquired
and saved. At the same time, a probe of the Moor oxygenation monitor (VMS-OXY, Moor Instruments Inc., Devon, UK)
was adhered to the beaker, which recorded the actual oxygen saturation of the blood phantom during calibration procedure.
4.1.2. Experiment result
Figure 3. (A)Blood phantom StO2 measured by multispectral algorithm versus that measured by Moor oxygenation monitor.
(B)& (C) Blood phantom at full oxygenation and full deoxygenation. (D)& (E) StO2 map of blood phantom at full
oxygenation and full deoxygenation
A
d
e
Proc. of SPIE Vol. 8937 893706-5
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
220
Q200
180Q-' 160.o>"
-a 140EF, 1200g 100,-,á 80g 60
, 40á)a 20
0
O Perfusion result
linear fitting
0 2 4 6 8White diffuser moving speed (trim's)
10
At each oxygenation levels of each recipe, StO2 were reconstructed from multispectral images of 544, 552, 568, 576, 592,
600nm using our algorithm as described previously. Three region of interest (ROIs) on the maps were selected to average
the StO2 values. After normalized to the scale of 0% to 100% by a linear function, the results were compared with the
actualStO2measured by Moor oxygenation monitor as plotted in Figure 3(A). According to the figure, linear correlation
can be observed between the calculated oxygenation levels and their actual values, indicating that the algorithm was able
to reconstruct tissue oxygenation. Figure 3(B) - (E) intuitively show the contrast of blood phantom at full oxygenation,
deoxygenation and corresponding StO2color maps.
4.2. Calibration experiment for laser speckle imaging
4.2.1. Calibration materials and methods
Here we introduce an integrated calibration method for accurate laser speckle imaging using a Flux Standard and a moving
white diffuser. The Flux Standard (Moor Instruments Inc., Devon, UK) was a colloidal suspension of polystyrene particles.
After shacked gently for 10 seconds and left to rest for 2 minutes, the suspension would be imaged with laser illumination.
At room temperature (25℃), the stationary background area will produce a contrast value of 1, and the Flux Standard will
produce a perfusion value of 250±5 Perfusion Units (PU). Thus, the parameters β and k described in formula (2) and (3)
can be determined. Besides, the white diffuser (The National Institute of Standards and Technology, Gaithersburg, MD)
was pushed by a motorized positioning system (TSA100, ZOLIX instruments, China) at a constant velocity to simulate
the movements of blood flow [10]. The measurements were taken over the regular velocity range from 0 to 10 mm/s, and
at each velocity plateau, 30 consecutive speckle images were acquired at 5 frames per second, then the perfusion P were
calculated by LASCA algorithm as above described.
4.2.2 Experiment result
Figure 4. Perfusion measured by LASCA versus white diffuser moving speed
Firstly, the perfusion result of a Flux Standard and static background was calculated by original LASCA algorithm. Three
ROIs were selected to average the contrast and perfusion values. According to Formula (2) and (3), to ensure C=1 for static
background and P=250±5 PU for Flux Standard, the parameters β and k were determined to be 2.00 and 21.45 respectively.
Perfusion (PU)
A
B
Proc. of SPIE Vol. 8937 893706-6
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
CCDcamera
light ring
AOTF light source
Moor oxygenation& perfusion monitor
electronic L
shutter 785nm Laserlight source
Tissue NIST traceablesimulating white diffuserphantom
Secondly, as shown in Figure 4(A), perfusion produced from the dynamic speckle on white diffuser increases with the
moving velocity linearly within a regular range of blood flow, the fitting line reveals an impressive correlation between
two parameters. Figure 4(B) shows four representative perfusion maps at different moving speeds. Combing this two
experiment results into account, the linearity and numerical accuracy of laser speckle imaging can be verified.
4.3. Test on a tissue-simulating and blood circulation phantom
4.3.1. Phantom Fabrication Protocol
In order to simulate the optical characteristics and blood circulation in cutaneous tissue, we designed a new style of solid-
state, biocompatible, permanent, and reusable phantom containing vessel-simulating channels. The phantom was
composed of polydimethylsiloxane (PDMS), Titanium dioxide powder (TiO2), and India ink. Firstly, we prepared PDMS
(Dow Corning, US) and matched hardener for matrix solidification in a ratio of 10:1 by weight. Secondly, designed dosage
of TiO2 (Guangfu Fine Chemical Research Institute, China) and India ink (Bomei Biotechnology Co., Ltd, Hefei, China)
were added into the matrix as the scattering and absorbing agents respectively. After stirred to homogeneity, the mixture
would be put in the vacuum pump for air exhaust. Then we poured the mixture into a rounded mold with a square section
vessel model at proper position and left it to harden for several hours. Finally, the vessel model was taken out when the
phantom was completely solidified. In order to check the phantom’s optical characteristics, an OxiplexTS tissue
spectrophotometer was used to confirm the absorption coefficients of PDMS phantom to be 0.122 cm-1 (at 690 nm) and
the reduced scattering coefficient to be 5.3 cm-1 (at 690 nm), the optical parameters was match with those of normal human
cutaneous tissue.
4.3.2. Experiments procedure and result
Figure 5. Schematic drawing of an experiment setup for simulating blood circulation and oxygenation adjustment
An experiment system as shown in Figure 5 was setup to simulate the blood circulation and oxygenation adjustment. Two
completely duplicate PDMS phantoms were connected in series by silicone tubes, one was used for multimodal imaging
and the other one was used to validate the imaging result via an affixed Moor oxygenation and perfusion detector (VMS-
OXY & VMS-LDF, Moor Instruments Inc., Devon, UK). To realize the simulating blood circulation in phantoms, a
Proc. of SPIE Vol. 8937 893706-7
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
Perfusion (PU) St0
peristaltic pump (BT100-2J, Baoding Longer Precision Pump Co., Ltd, China) was used to drive 250 mL blood solution
(diluted by distilled water in a ratio of 1:5 by volume) sealed in a conical flask, which could eliminate the effect of oxygen
in ambient air and weaken the intermittency of peristaltic pump. The oxygenation level of the blood was adjusted by
dropwise addition of sodium hydrosulfite (0.025g/mL) using an injection syringe through a silicone tube. At each
oxygenation plateau, the blood phantom was mixed to homogeneity by a magnetic stirrer. On the basis of this system, the
multispectral imaging and laser speckle imaging can be verified on PDMS phantom simultaneously or respectively.
Figure 6. (A)& (B) PDMS phantoms used for imaging and verification. (C)& (D) Perfusion maps of the phantom at
stationary blood flow and high blood flow. (E)& (F) Oxygenation maps of the phantom at full oxygenation and full
deoxygenation.
Figure 6(A) & (B) demonstrated the RGB images of two tissue-simulating phantoms and the probe of Moor oxygenation
and perfusion monitor. When the peristaltic pump was laid off, the simulating blood flow was static and perfusion map
showed a low level as Figure 6(C), meanwhile the Moor perfusion monitor indicated to be 5±3 PU. As a comparison,
Figure 6(D) exhibited the perfusion map of high blood flow in the vessel simulating channels, when the Moor monitor
indicated to be 100±10 PU. With regard to the oxygen saturation adjustment, StO2 maps of the phantom at full
oxygenation and full deoxygenation were shown in Figure 6(D) & (E), accordingly, the Moor oxygenation monitor
displayed as 56.7% and 0%. The elementary test results revealed the potential of this new style of phantom to verify our
multimodal imaging system.
5. IN VIVO VERIFICATION ON CUTANEOUS TISSUE
5.1. Human skin StO2 and perfusion dynamics during post-occlusive reactive hyperemia (PORH)
The multimodal imaging system of measuring StO2 and perfusion was verified on a healthy volunteer following a protocol
of post-occlusive reactive hyperemia (PORH) [11]. The subject’s hand was comfortably rested on a countertop within the
view field of the CCD camera. A Moor tissue oxygenation and perfusion detector was affixed next to the monitoring region.
The PORH process consisted of a pre-occlusive baseline period (no pressure applied to the arm) of one minute, a
suprasystolic occlusion (180 mm Hg) period of three minutes, and a reactive hyperemia period (pressure released) of two
minutes. During the PORH test, the illumination was switched from multispectral light beam to laser light beam by
electronic shutter periodically, correspondingly, multispectral images of six designated wavelengths (544, 552, 568, 576,
592, 600nm) and laser speckle images were acquired. It took about 6 to 8 seconds to finish one acquisition period. At the
meantime, Moor oxygenation and perfusion monitor recorded the dynamics of oxygenation and perfusion during the
PORH process.
Proc. of SPIE Vol. 8937 893706-8
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
Blo
od P
erfu
sion
PU
can
SoI
I
LO
tg
Oxy
gen
Sat
urat
ion
A
Teflon Diffuse
Figure 7. Comparison of the measurements results between the imaging system and Moor tissue oxygenation and laser
Doppler monitor during a PORH process
After image acquisition, tissue oxygen saturation and perfusion parameters of the monitoring region were reconstructed
by the calibrated algorithms as described previously. Three ROIs on the monitoring region were selected to even out the
possible errors. Figure 7 shows the comparison of the measurements results between multimodal imaging system and the
Moor oxygenation and perfusion monitor during the PORH process. It can be seen from the green curves that the StO2
measurement of both the imaging system and the Moor monitor matches well. The StO2 dropped during suprasystolic
occlusion period (60s ~ 240s) and shot up above the starting StO2 when the pressure on the arm was released (240s ~ 360s).
Similar trend of perfusion measurement is plotted by the blue curves, but there still existed some disparities between the
two measurements in the initial and final stages. The difference may due to two reasons. On one hand, the acquisition
period of multimodal images took about 6-8 seconds due to the responsive time of electronic shutter switching, enough
exposure time for capturing every image, and the time delay of Labview Run-Time software package. Thus, optimization
of the software and reducing the time required for illumination switch will improve the accuracy of the measurement. On
the other hand, the measurement position and depth of Moor monitor were not exactly coincide with the imaging system.
Besides, the Moor monitor maybe affected by the variation of the blood concentration, skin color and scattering properties
of the human subject.
5.2. Wound healing process monitoring using dorsal skinfold chamber models
5.2.1. Dorsal skinfold chamber model setup
Figure 8. (A) Nude mouse implanted of a dorsal skin-fold chamber; (B) a modified mice holder; (C) Dorsal skin-fold wound
chamber model with a Teflon diffuser
Proc. of SPIE Vol. 8937 893706-9
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
In this study we established in vivo mouse dorsal skin-fold chamber models in order to monitor the dynamics of
oxygenation and perfusion during cutaneous wound healing process. Four male BALB/c-nu nude mice (Vital River
Laboratory Animal Technology Co. Ltd., China) of 10 weeks old with a body weight of 23-25g were used for the
experiments. Figure 8(A) shows a nude mouse implanted of a dorsal skin-fold chamber, the fold of the depilated dorsal
skin of a mouse was taken and then fixed like a sandwich between the two titanium frames of the chamber. A full dermal
thickness wound of 2mm in diameter was made by removing the skin (including epidermis, dermis, subcutis, cutaneous
muscle, and subcutaneous fatty tissue) of one skin layer completely using a biopsy punch. A sterile coverslip was used to
close the operation filed to avoid infection or mechanical damage. To make it convenient for multimodal assessment by
imaging system, a modified mice holder was developed to fix the mouse as show in Figure 8(B). Furthermore, a calibrated
white diffuser made of Teflon was used to replace the NIST 99% reflected diffuser for multispectral imaging as shown in
Figure 8(C).The experiments were conducted in accordance with guidelines for the Care and Use of Laboratory Animals
and the Institutional Animal Care and Use Committee of University of Science and Technology of China.
5.2.2. Experiment results
Figure 9. Dynamics maps of oxygen saturation and blood perfusion of tissue in wound healing process
For the sake of monitoring the dynamic tendency of wound tissue functional characteristics during the healing process, the
sequential multimodal images were captured from the first day after biopsy to the ninth day, when the wound had
remodeled completely. Oxygen saturation and blood perfusion maps of the wound tissue areas were reconstructed and
compared. Figure 9 enumerates the representative result series of wound tissue StO2 and perfusion dynamics during the
wound healing process. Considering the StO2 series, tissue oxygen saturation at the first day indicated to be a relative low
level. However, an obvious increase was witnessed from day 2 to day 5 after biopsy. After that, the StO2subsided slowly
and reached a stable oxygen level as healing completed. Towards the perfusion, similar dynamic trend could be seen except
that the peak blood flow of wound tissue occurred later in contrast with the rising of the StO2. It could be deduced that the
initial dip in StO2 and perfusion may be attributed to the hemostasis due to biopsy. And with the growth of vessels at 3-6
days, the parameters exhibited a significant increase, the results were consistent with our current understanding of the
positive and negative regulation of angiogenesis during the different phases of wound healing [12].
Days 1 2 3 4 5 6 7 8 9
Wound
StO2
Perfusion
Proc. of SPIE Vol. 8937 893706-10
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
6. CONCLUSION
In this paper, we present a simultaneous and multimodal imaging system that integrates multispectral imaging module and
laser speckle imaging module. The system was capable of quasi-simultaneous and quantitative characterization of tissue
oxygenation and perfusion in the same region, which was significant to assess wound healing. To calibrate and verify the
imaging system, two quantitative calibration experiments and a new style of skin-simulating phantoms were designed,
therefore the numerical accuracy and reliability of the imaging system can be assured. In vivo, a PORH procedure test on
human hand presented commendable results comparing with a Moor oxygenation and perfusion monitor. What’s more, an
ongoing wound healing monitoring experiment using nude mice dorsal skinfold chamber models revealed the dynamic
tendency of StO2 and perfusion parameters during the healing process. All these results demonstrate the potential of the
imaging system for cutaneous tissue assessment and diagnosis in clinical practice. The future work will focus on enhancing
the accuracy and processing speed of the system and integrating structural and functional characteristics detection together.
ACKONWLEDGEMENTS
The project was partially supported by Natural Science Foundation of China (Nos. 81271527 and 81327803) and the
Fundamental Research Funds for the Central Universities (WK2090090013).
REFFERENCES
[1] Sen, Chandan K., et al. "Human skin wounds: a major and snowballing threat to public health and the economy,"
Wound Repair and Regeneration, 17(6), 763-771 (2009).
[2] Broughton 2nd, G., Jeffrey E. Janis, and Christopher E. Attinger. "The basic science of wound healing," Plastic
and reconstructive surgery, 117(7), 12S-34S (2006).
[3] Sen, Chandan K. "Wound healing essentials: let there be oxygen," Wound Repair and Regeneration, 17(1), 1-
18(2009).
[4] Singer, Adam J., and R. A. Clark. "Cutaneous wound healing," N Engl J Med, 341(10), 738-746, (1999).
[5] Brian W. Pogue., Micheal S. Patterson., "Review of tissue simulating phantoms for optical spectroscopy, imaging
and dosimetry," J Biomed Opt, 11(4), 041102 (2006).
[6] S. Prahl, "Simple and Accurate Approximations for Reflectance from a Semi-Infinite Turbid Medium," OSA
Biomedical Topical Meetings, 613-615 (2002).
[7] D. E. Myers, L. D. Anderson, et al., "Noninvasive method for measuring local hemoglobin oxygen saturation in
tissue using wide gap second derivative near-infrared spectroscopy," J Biomed Opt, 10(3), 034017-03401718
(2005).
[8] Briers, J. D. & Webster, S. "Laser speckle contrast analysis (LASCA): a nonscanning, full-field technique for
monitoring capillary blood flow," J Biomed Opt, 1(2), 174-179 (1996).
[9] Ronald X. Xu, et al. "Developing digital tissue phantoms for hyperspectral imaging of ischemic wounds,"
Biomedical Optics Express, 3(6), 1433 (2012).
[10] Li Pengcheng, et al. "Imaging cerebral blood flow through the intact rat skull with temporal laser speckle
imaging," Optics Letters, 31(12), 1824-1826 (2006).
[11] Arkin, H., L. X. Xu, and K. R. Holmes. "Recent developments in modeling heat transfer in blood perfused
tissues," Biomedical Engineering, 41(2), 97-107 (1994).
Proc. of SPIE Vol. 8937 893706-11
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx
[12] Rege, Abhishek, et al. "In vivo laser speckle imaging reveals microvascular remodeling and hemodynamic
changes during wound healing angiogenesis," Angiogenesis, 15(1), 87-98 (2012).
Proc. of SPIE Vol. 8937 893706-12
Downloaded From: http://proceedings.spiedigitallibrary.org/ on 09/30/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx