Journal of Medical Engineering & Technology
ISSN: 0309-1902 (Print) 1464-522X (Online) Journal homepage: http://www.tandfonline.com/loi/ijmt20
Characteristics of laser Doppler flowmeters with differing optical arrangements K. Mito To cite this article: K. Mito (1992) Characteristics of laser Doppler flowmeters with differing optical arrangements, Journal of Medical Engineering & Technology, 16:6, 236-242, DOI: 10.3109/03091909209030774 To link to this article: http://dx.doi.org/10.3109/03091909209030774
Published online: 09 Jul 2009.
Submit your article to this journal
Article views: 2
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ijmt20 Download by: [RMIT University Library]
Date: 22 May 2016, At: 05:34
Journal of Medical Engineering & Technology, Volume 16, Number 6 (November/December 1992), pages 256242
Characteristics of laser Doppler flowmeters _ _ with differing optical arrangements K. Mito
The LDF systems examined, as shown in figure 1, were: Periflux (Perimed Co. Ltd, Sweden): He-Ne laser, 632.8 nm, 2 mW; ALF, 21 (Advance Co. Ltd, Japan): semiconductor laser, 780 nm, 2 mW; Diodopp (Applied Laser Technology BV, The Netherlands): semiconductor laser, 780 nm, 2mW.
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Kawasaki Collele of Allied Health Professions, Matsushima 316, Kurashiki 701-01,Japan
Recently, various laser Doppler flowmeter ( L D F ) systems have been used f o r evaluation of tissue microcirculation. However, indifference has often been shown by doctors and researchers regarding their sampling volume, stability and accuracy, which dffer with various optical arrangements. In this study we evaluated the characteristics of three LDF systems with different optical arrangements. The sampling depths in both blood and tissue werefound to differ with the different optical arrangements, although the laser power emitted from the probes is nearly equivalent. Each depth signijicantly changed at haematocrits between 5.0 and 36.5%. The stability and the accuracy of the measurements also difjred. In the practical case of tissue microcirculation it is important to measure blood Jlow with an understanding of the material differences in the characteristics of the L D F systems.
In the Periflux system the light is transmitted to the sampling point on the skin by optical fibres connected to the He-Ne laser tube. The scattered light is received by two other optical fibres and transmitted back to the detector. The interval, L, which is the distance between the centre of the emitting and receiving fibres, is 0.82 mm. I n the ALF-21 system the light emitted from the diode is transmitted to the skin by an optical fibre and received by another optical fibre; L is 0.66 mm. In the Diodopp system the diode and two photo detectors are included in the probe so there are no optical fibres; L is 2.73 mm, which is the greatest distance between emitting point and receiving point in the three systems.
Introduction
Methods
Laser Doppler flowmetry (LDF) is an effective and reliable method for the evaluation of tissue microcirculation. Recently, semiconductor laser diodes with nearinfrared light have been used in LDF systems as a light source instead of He-Ne laser tubes, because they are smaller, cheaper and generally more stable. Although the optical parameters of tissue may depend on the particular light source used, doctors and researchers are often indifferent to the sampling volume, stability and accuracy of the LDF they are using. The sampling depth is especially important, because the longer wavelength of near-infrared light in diodes is known to penetrate more deeply into tissue, and has different optical parameters in cutaneous tissue than light from an He-Ne laser. Obeid et al. compared the relative responses of a near-infrared laser diode system (wavelength 780 nm) with those of an He-Ne laser system (632.8 nm) in both an in vitro model and animal tissue. They suggested that the larger the wavelength the greater the volume that can be sampled [1,2], but they did not report any specific difference in the volume sampled.
Evaluation of sampling depth The shape of the sampling volume of the LDF is assumed to be a hemisphere in the vicinity of the probe terminal, as shown in figure 2. It is difficult to evaluate the sampling depth of the volume in living tissue. For this reason the depth was evaluated comparatively from the maximum detectable depth obtained in both canine blood and porcine tissue in vitro. The maximum detectable depth (dmaX) was defined as the depth at which the Doppler signal disappeared as the distance (d)
If accurate evaluation of living tissue microcirculation in both clinical medicine and microvascular research is to be achieved, the blood flow must be measured with knowledge of the material characteristics of LDF systems using different optical arrangements. In this study the sampling volume, stability and accuracy in blood flow measurement of three LDF systems were evaluated and compared.
Figure I . Three L D F systems: (a) PeriJux, (b) ALF-21, and (c) Diodopp.
236 0309-1902/92 fl0.M)
0 1992 Taylor & Francis Ltd.
K. Mito Characteristics of laser Doppler Rowmeters with differing optical arrangements
P Probe
Depth)
Sampling Volume Figure2. The assumed shape of the sampling volume of the L D F .
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Reservoir
Figure 3. Schematic conjiguration of a blood Jlow model for the evaluation of the maximum detectable depth of blood b_r L D F . Blood was sprayed onto the glass plate attached to tube B to obtain a Doppler signal. The depth d was changed step by step using a micromanipulator. between the probe and flowing blood was increased. In practice the measurable maximum depth in living tissue will be less than this value. The maximum detectable depth in blood was evaluated using the system shown in figure 3. It consisted of acrylic tubes A and B, and a glass capillary C. The glass capillary was connected to a reservoir, and tube B was connected to a collecting device D. Two thin transparent glass plates (thickness 0.13-0.16 mm, Matsunami Glass Ind., Ltd, Japan) were attached to the bottoms of the two tubes. Canine whole blood was heparinized and diluted in physiological saline. The haematocrit of the blood was prepared at both 5.0 and 36.5% (whole blood). Blood was stored in tube A and frequently stirred to avoid sedimentation. The blood which was to be measured was perfused constantly from the reservoir to the collecting device D via the capillary C. Then this blood was sprayed onto the inside wall of the glass plate of tube B to provide the Doppler signal. The light was emitted from the outer side of the glass plate of tube A to the sprayed blood through the stored blood in the
tube A. The depth d , the distance between the bottom of tube B and the bottom of tube A, was varied step by step using a micromanipulator. The intensity of the Doppler signal was measured at each step. The maximum detectable depth in tissue was evaluated using porcine tissue. The experimental system, which consisted of a flow channel A and a chamber B, is shown in figure 4. The channel was a rectangular flow channel (width 24 mm) and its upper and lower sides consisted of transparent glass plates. The upper plate was the lower plate of the chamber B which with another plate D formed a wedge-shaped chamber. The two plates were made of thin glass (0.13-0.16 mm). The tissue sample was removed from porcine muscle, carefully washed in physiological saline, frozen and cut in the shape of a wedge. After cutting, the tissue was placed in the chamber and gradually thawed at room temperature to fix it smoothly on both plates. The thickness of the widest part of the chamber was 6 mm. The probe of the LDF was attached to the outside of the upper thin glass plate D and traversed along the slope from the narrowest to the widest part of the tissue. Therefore, the thickness of the porcine tissue changed 237
K. Mito Characteristics of laser Doppler flowmeters with dilfering optical arrangements
Plate (D)
Reservoir Chamber (B)
Flow
I I +-romm-
kl
II
Flowing Blood
Collecting Device
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Figure 4. Schematic configuration of the arrangement f o r the evaluation of the maximum detectable depth of tissue. The chamber (B) was formed in the shape of a wedge of tissue. The Doppler signal w a s obtained from the bloodjowing in a rectangularjow channel ( A ) beneath the tissue wedge.
* White Paper
1Omm 20mm
/
-JL f
*
.
1
Turntable
f
Motor
-k
300mm-4
Figure5. Rotating turntable f o r the evaluation o f the stability of LDF. The Doppler signal was obtained f r o m the surface of the white paper. from zero to about 6 mm. The laser light was passed to the flow channel through the tissue. T o obtain the Doppler signal the blood was perfused constantly from the reservoir to the collecting device via the flow channel. The intensity of the Doppler signal was measured at various tissue thicknesses.
Signal jluctuation Signal fluctuation, which affects flow measurement, was investigated using a rotating turntable as shown in figure 5. A sheet of white paper was placed on the turntable, and it was rotated at a constant number of revolutions. The probe was placed at a distance of approximately 10 mm from the surface of the paper. The Doppler signal patterns were sequentially recorded by each LDF system.
Accuracy of blood jlow measurement Since it is generally difficult to measure the absolute blood flow velocity in living tissue, it was evaluated using glass capillary tubes as shown in figure 6. The experimental system consisted of a distribution reservoir A, a tube B, a measuring section C, an electromagnetic flowmeter EMF (Nihonkoden Co. Ltd, Japan) and a collecting device D. The inlet of the measuring section was connected to the reservoir and its outlet was 238
connected to the collecting device. The inner diameter of the tube at the EMF was 8.0 mm. The measuring section (figure 6b) consisted of 89 parallel glass capillary tubes (inner diameter 0.2 mm). This was similar to the mechanical fluid model reported by Nilsson et al. [3], which consisted of 37 symmetrically-arranged tubes (inner diameter 0.3 mm) and a 0.2-mm thick slot in a radial direction toward the periphery between the edge of the tubes and the optical sensor. In the present system a wider measuring section was needed to receive the illumination of the light in the Diodopp system, in which the distance between the laser diode and photodetector is relatively large (L=2*73 mm). Consequently, it was necessary to use 89 tubes. Thc distance between the end of the capillaries and the bottom of the tube B, which was the flow channel for flow from the capillaries, was 2.5 mm. The haematocrit of the perfused blood was 12%. Each LDF probe was successively attached to the outer side of the transparent glass plate at the bottom of the tube B (figure 6b). The blood flow was measured at each flow volume, as the blood was perfused at a constant rate.
Results Evaluation of the sampling depth As the sampling depth in blood is affected by the power of the emitted light, the power of the light of each
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements
Table 1. Emitted laser power of each LDF system
Table 2. Maximum detectable depth in blood usins each LDF system in the cases of haematocrit 5.0 and 36.5% (whole blood)
Emitted laser power (mW)
LDF system
1.98 1.99 - 2.02 1.99
Periflux ALF-2 1 Diodopp
LDF system
Haematocrit 5.0°/0
Haematocrit 36.5%
2.0
02
4.0 3.0
2.6 2.0
Periflux ALF-2 1 Diodopp
Downloaded by [RMIT University Library] at 05:34 22 May 2016
system was previously measured with an optical power meter (Model 44XLA, Photodyne Inc., USA). The power of all three systems was confirmed to be nearly equal (= 2 mW), but the stability of the ALF-21 was poor compared to that of the other systems, as shown in table 1.
in which a He-Ne laser (632.8 nm) is used. Compared to the Periflux, the ALF-21 was 2.0 times more sensitive with the lower haematocrit blood and the Diodopp was 1.5 times more sensitive. With the higher haematocrit blood, the value for the ALF-21 was one order of magnitude greater than the Periflux value.
The Doppler signal gradually disappeared as the depth of stored blood was increased. The relationship of the signal intensity to the depth was exponential, as shown in figure 7. The maximum detectable depth is shown by the arrow in each figure, and is shown in table 2 for haematocrits 5.0 and 36.5%. I n both cases the maximum depth was obtained with the ALF-21, followed by the Diodopp system; in both systems a longwavelength (780 nm) semiconductor laser diode is used. The minimum depth was obtained with the Periflux,
With regard to tissue, the relationship of the signal intensity to the depth is shown in figure 8. The maximum depth (dmaX)reached by each system is given in table 3, and indicated by the arrows in figure 8. The maximum depth in the case of blood was obtained with the ALF-21 followed by the Diodopp and Periflux systems, respectively. Compared with the Periflux, the
Reservoir (A)
(a)
7
I
0 Blood
f E
E
0
ow Chamber (6) easuring Section (C)
(0
I
ass Capillary Tubes)
' II
Collecting Device (D)
C30mm-4
Figure 6. (a) Schematic conjguration of the experimental arrangement for determining the accuracy of blood flow measurement. Blood was perfused through the measuring section (C) which was composed of 89 capillary tubes. The laser beam was emitted at the edge of the assembled capillary tubes to obtain the Doppler signal. (b) Enlarged illustration of the capillary tube measuring section.
05-
10
I --.
g
..__
20
30
40
Thickness of Blood Layer (rnrn)
(a)
Thickness of Blood Layer (rnrn)
(b)
Thickness of Blood Layer (rnrn) (C)
Figure 7. Relationship between the detected Doppler signal intensity and the thickness of stored blood (d) in the case of blood haematocrit 5.0%. The signal intensity is given in arbitrary units (A U).The arrows indicate the maximum detectable thickness d,,,. (a) Periyux, (a) ALF-21, (c) Diodopp. 239
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements
1 .o
3
5
0
a a 0
.Q 0
10
20
.. 30
.
0
' e 9 .
0 40
50
60
Thickness of Blood Tissue (mm)
10
20
t 2 .... 40
50
C
10
60
Thickness of Blood Tissue (mm)
(a)
Downloaded by [RMIT University Library] at 05:34 22 May 2016
30
.. *.p. .... 20
30
40
50
60
Thickness of Blood Tissue (mm)
(b)
(C)
Figure 8. Relationship between the detected Doppler signal intensity and the thickness of tissue. The signal intensity is given in arbitrary units ( A U). The arrows indicate the maximum detectable thickness d,,, (a) PeriJux, (b) ALF-PI, (c) Diodopp.
M 1 min
Time
Time
1 mi"
Figure 9. The jluctuation of the Doppler signal when the laser probe is placed perpendicular to a surface of a moving white paper.
ALF-21 was about 2.3 times more sensitive and the Diodopp was about 1.9 times more sensitive. Signal jluctuation The Doppler signal patterns sequentially obtained by each system are shown in figure 9. Small periodic fluctuations of the signal coinciding with the period of the rotational speed of the turntable were observed in each pattern. As this fluctuation was considered to be due to a slight distortion of the paper surface, signals Table 3. Maximum detectable depth in tissue usins each LDF system
Periflux ALF-2 1 Diodopp ~~
240
1.5 3.5 2.8
were sampled for the same temporal period. The coefficient of variation (c.v.) of each fluctuation was 0.53 in the Periflux, 0.81 in the ALF-21 and 0.27 in the Diodopp. The Diodopp system showed the highest stability, and the ALF-21 shows the lowest stability. The lower stability of the ALF-21 may be due to instability of the laser source, as shown in table 1 .
Accuracy of blood Jlow measurement The Doppler signals obtained are shown in figure 10. Arbitrary units were used, and they were standardized by the maximum value in each case. The Periflux and Diodopp systems showed good linearities between the Doppler signal and blood flow volume. However, the linearity between the Doppler signal and blood volume for the ALF-21 was poor. It may be due to the difference of the sampling depth in the flow channel.
K. Mito Charactrristics of lasrr Doppler flowmeters with direring optical arrangements
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Discussion The maximum detectable depth differed for each optical arrangement, although the power of the emitted light from the three systems is nearly equivalent. The maximum value in blood and tissue was obtained with the ALF-2 1, followed by the Diodopp and the Periflux (figure 11). It was confirmed that a LDF system with a long-wavelength semiconductor laser diode (780 nm) can sample at a greater depth than those with an He-Ne laser (632.8 nm). Our results show that differences exist in the optical parameters of both blood and tissue, i.e. an absorbing coefficient, a scattering coefficient and a phase function. Since Van Gemert et al. [ 4 ] suggested that the absorbing and scattering coefficients at a wavelength of 632.8 nm are similar to those at a wavelength of 780 nm, it was considered that the differences were mainly the result of differences in the phase function at short (632.8 nm) and long wavelengths (780 nm).
for blood. As the tissue structure in the sampling volume is probably unchanged, it is assumed that the sampling depth in the practical tissue microcirculation case is affected by the changeable haematocrit in several disease conditions respectively. The variation in the depth obtained was the largest in the Periflux, followed by ALF-21, and the Diodopp, as shown in Figure 11. It is, therefore, suggested that the Diodopp system should be used as the change of haematocrit has the smallest effect in practical blood flow measurement. Jentink et al. [5] reported that the sampling volume of LDF can be extended to deep layers by increasing the distance (L) between the laser beam and detector using the Monte-Carlo simulation. This distance is 0-82 mm in the Periflux system, 0.66 mm in the ALF-21 and 2.73 mm in the Diodopp. However, the maximum detectable depth was obtained with the ALF-21 with the minimum distance L and not with the Diodopp. Therefore, the depth may be affected not only by the distance but also by the numerical aperture.
Figure 11 suggests clearly that the sampling depth in the practical tissue microcirculation case will differ if the optical arrangements of the LDF systems differ. Furthermore, the depth of the measurement of each system will vary with haematocrit. In this study the depth obtained in tissue was within the range of that
With regard to signal fluctuation, the ALF-21 and Periflux systems were not stable. Periodic noise was observed in the signal of the Periflux system. This may be the laser cavity warming-up effect, as reported by Obeid et al. [ 6 ] . The ALF-21 shows instability which
. ..
s
f
0
50
1.0
4
*
...I
100
Blood Flow Volume (EMF) (ndlmin)
Blood Flow Volume (EMF) (nllmin)
Blood Flow Volume (EMF) (nl/min)
(a)
(b)
(C)
Figure 10. Relationship belmeen the Doppler shiJ signal and [he known blood flow volume: (a) PeriJux, (b) ALF-21, (c) Diodopp.
5n
rnrn 4.0
-
40
d Q
0 Tissue
3.0-
3.0
c 0 0 c Q
n
5E
2.6
2.0 -
2.0
.X
i?-
1.0-
0.2
-I
0.’
Periflux
ALF- 21
2.0
Diodopp
Figure 11. The maximum delectable depth in blood and tissue. 24 1
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements may be due to fluctuation of the power of the emitted laser light, as shown in table 1.
Downloaded by [RMIT University Library] at 05:34 22 May 2016
As for the accuracy of blood flow measurements, the Diodopp and Periflux systems showed good linearities between Doppler signals and blood flow volumes, but the ALF-21 did not. This may be due to difference in the sampling depth: in the ALF-21, many blood flow measurements, including the flow in the capillary tubes, were obtained in sprayed blood between the edge of the tubes and the inside glass plate, as shown in figure 3, with its large sampling volume.
Conclusion In this study, sampling depth, stability and accuracy clearly differ with different optical arrangements. Although the data obtained in this study are only relative, they provide important information regarding LDF systems with different optical arrangements. In measuring blood flow in the practical tissue microcirculation case, a moderate sampling depth, smaller variance with the change in haematocrit, and stability are necessary for accurate measurement. For this purpose a system which can be directly applied to blood flow using a near-infrared laser light without optical fibres, such as the Diodopp system, seems most promising.
242
Acknowledgement
I express my thanks to M r Osamu Yasuda, General Manager (Nikko Keisoku System Ltd) and Mr Nobuhiko Ueda, Ms Tomoko Fujita (Nikko Keisoku System Ltd), Fumika Nakamura, Shinichi Miyake and Katsuya Tsuda (Kawasaki College of Allied Health Professions) for their cooperation in the experiment. References 1. OBEID,A. N., BOGCETT, D., DOUGHERTY, G., BARNETT, N. J. and ROLGE, P. (1988) Depth discrimination in laser
Doppler skin blood flow measurement using different lasers. Medical and Biological Engineering and Computing, 26, 415-419. 2. OBEID,A. N., DOUGHERTY, G. and PETTINCER, S. (1990) In vivo comparison of a twin wavelength laser Doppler flowmeter using He-Ne and laser diode sources. Journal of Medical Engineering and Technology, 14, 102-1 10. 3. NILSSON, G. E., TENLAND, T. and OBERG,P. (1980) Evaluation of a laser Doppler flowmeter for measurement of tissue blood flow. I E E E Transactions on Biomedical Engineering, BME-27,597-604. 4. VANGEMERT, M. J. C., JACQUES, S. L., STERENBORG, H. J. C. M. and STAR,W. M. (1989) Skin optics. IEEE Transactions on Biomedical Engineering, 56, 1146-1 154. 5. JENTINK, F. F., DE MUL, M., HERMSEN, R. G. A. M., GRAAFF, R. and GREVE, J. (1990) Monte Carlo simulations of laser Doppler blood flow measurements in tissue. Applied Optics, 29, 2371-2381. 6 . OBEID,A. N., BARNETT,N. J., DOUGHERTY, G. and WARD,G. (1990) A critical review of laser Doppler flowmetry. Journal of Medical Engineering and Technbiogy, 14, 178-181.
ISSN: 0309-1902 (Print) 1464-522X (Online) Journal homepage: http://www.tandfonline.com/loi/ijmt20
Characteristics of laser Doppler flowmeters with differing optical arrangements K. Mito To cite this article: K. Mito (1992) Characteristics of laser Doppler flowmeters with differing optical arrangements, Journal of Medical Engineering & Technology, 16:6, 236-242, DOI: 10.3109/03091909209030774 To link to this article: http://dx.doi.org/10.3109/03091909209030774
Published online: 09 Jul 2009.
Submit your article to this journal
Article views: 2
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ijmt20 Download by: [RMIT University Library]
Date: 22 May 2016, At: 05:34
Journal of Medical Engineering & Technology, Volume 16, Number 6 (November/December 1992), pages 256242
Characteristics of laser Doppler flowmeters _ _ with differing optical arrangements K. Mito
The LDF systems examined, as shown in figure 1, were: Periflux (Perimed Co. Ltd, Sweden): He-Ne laser, 632.8 nm, 2 mW; ALF, 21 (Advance Co. Ltd, Japan): semiconductor laser, 780 nm, 2 mW; Diodopp (Applied Laser Technology BV, The Netherlands): semiconductor laser, 780 nm, 2mW.
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Kawasaki Collele of Allied Health Professions, Matsushima 316, Kurashiki 701-01,Japan
Recently, various laser Doppler flowmeter ( L D F ) systems have been used f o r evaluation of tissue microcirculation. However, indifference has often been shown by doctors and researchers regarding their sampling volume, stability and accuracy, which dffer with various optical arrangements. In this study we evaluated the characteristics of three LDF systems with different optical arrangements. The sampling depths in both blood and tissue werefound to differ with the different optical arrangements, although the laser power emitted from the probes is nearly equivalent. Each depth signijicantly changed at haematocrits between 5.0 and 36.5%. The stability and the accuracy of the measurements also difjred. In the practical case of tissue microcirculation it is important to measure blood Jlow with an understanding of the material differences in the characteristics of the L D F systems.
In the Periflux system the light is transmitted to the sampling point on the skin by optical fibres connected to the He-Ne laser tube. The scattered light is received by two other optical fibres and transmitted back to the detector. The interval, L, which is the distance between the centre of the emitting and receiving fibres, is 0.82 mm. I n the ALF-21 system the light emitted from the diode is transmitted to the skin by an optical fibre and received by another optical fibre; L is 0.66 mm. In the Diodopp system the diode and two photo detectors are included in the probe so there are no optical fibres; L is 2.73 mm, which is the greatest distance between emitting point and receiving point in the three systems.
Introduction
Methods
Laser Doppler flowmetry (LDF) is an effective and reliable method for the evaluation of tissue microcirculation. Recently, semiconductor laser diodes with nearinfrared light have been used in LDF systems as a light source instead of He-Ne laser tubes, because they are smaller, cheaper and generally more stable. Although the optical parameters of tissue may depend on the particular light source used, doctors and researchers are often indifferent to the sampling volume, stability and accuracy of the LDF they are using. The sampling depth is especially important, because the longer wavelength of near-infrared light in diodes is known to penetrate more deeply into tissue, and has different optical parameters in cutaneous tissue than light from an He-Ne laser. Obeid et al. compared the relative responses of a near-infrared laser diode system (wavelength 780 nm) with those of an He-Ne laser system (632.8 nm) in both an in vitro model and animal tissue. They suggested that the larger the wavelength the greater the volume that can be sampled [1,2], but they did not report any specific difference in the volume sampled.
Evaluation of sampling depth The shape of the sampling volume of the LDF is assumed to be a hemisphere in the vicinity of the probe terminal, as shown in figure 2. It is difficult to evaluate the sampling depth of the volume in living tissue. For this reason the depth was evaluated comparatively from the maximum detectable depth obtained in both canine blood and porcine tissue in vitro. The maximum detectable depth (dmaX) was defined as the depth at which the Doppler signal disappeared as the distance (d)
If accurate evaluation of living tissue microcirculation in both clinical medicine and microvascular research is to be achieved, the blood flow must be measured with knowledge of the material characteristics of LDF systems using different optical arrangements. In this study the sampling volume, stability and accuracy in blood flow measurement of three LDF systems were evaluated and compared.
Figure I . Three L D F systems: (a) PeriJux, (b) ALF-21, and (c) Diodopp.
236 0309-1902/92 fl0.M)
0 1992 Taylor & Francis Ltd.
K. Mito Characteristics of laser Doppler Rowmeters with differing optical arrangements
P Probe
Depth)
Sampling Volume Figure2. The assumed shape of the sampling volume of the L D F .
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Reservoir
Figure 3. Schematic conjiguration of a blood Jlow model for the evaluation of the maximum detectable depth of blood b_r L D F . Blood was sprayed onto the glass plate attached to tube B to obtain a Doppler signal. The depth d was changed step by step using a micromanipulator. between the probe and flowing blood was increased. In practice the measurable maximum depth in living tissue will be less than this value. The maximum detectable depth in blood was evaluated using the system shown in figure 3. It consisted of acrylic tubes A and B, and a glass capillary C. The glass capillary was connected to a reservoir, and tube B was connected to a collecting device D. Two thin transparent glass plates (thickness 0.13-0.16 mm, Matsunami Glass Ind., Ltd, Japan) were attached to the bottoms of the two tubes. Canine whole blood was heparinized and diluted in physiological saline. The haematocrit of the blood was prepared at both 5.0 and 36.5% (whole blood). Blood was stored in tube A and frequently stirred to avoid sedimentation. The blood which was to be measured was perfused constantly from the reservoir to the collecting device D via the capillary C. Then this blood was sprayed onto the inside wall of the glass plate of tube B to provide the Doppler signal. The light was emitted from the outer side of the glass plate of tube A to the sprayed blood through the stored blood in the
tube A. The depth d , the distance between the bottom of tube B and the bottom of tube A, was varied step by step using a micromanipulator. The intensity of the Doppler signal was measured at each step. The maximum detectable depth in tissue was evaluated using porcine tissue. The experimental system, which consisted of a flow channel A and a chamber B, is shown in figure 4. The channel was a rectangular flow channel (width 24 mm) and its upper and lower sides consisted of transparent glass plates. The upper plate was the lower plate of the chamber B which with another plate D formed a wedge-shaped chamber. The two plates were made of thin glass (0.13-0.16 mm). The tissue sample was removed from porcine muscle, carefully washed in physiological saline, frozen and cut in the shape of a wedge. After cutting, the tissue was placed in the chamber and gradually thawed at room temperature to fix it smoothly on both plates. The thickness of the widest part of the chamber was 6 mm. The probe of the LDF was attached to the outside of the upper thin glass plate D and traversed along the slope from the narrowest to the widest part of the tissue. Therefore, the thickness of the porcine tissue changed 237
K. Mito Characteristics of laser Doppler flowmeters with dilfering optical arrangements
Plate (D)
Reservoir Chamber (B)
Flow
I I +-romm-
kl
II
Flowing Blood
Collecting Device
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Figure 4. Schematic configuration of the arrangement f o r the evaluation of the maximum detectable depth of tissue. The chamber (B) was formed in the shape of a wedge of tissue. The Doppler signal w a s obtained from the bloodjowing in a rectangularjow channel ( A ) beneath the tissue wedge.
* White Paper
1Omm 20mm
/
-JL f
*
.
1
Turntable
f
Motor
-k
300mm-4
Figure5. Rotating turntable f o r the evaluation o f the stability of LDF. The Doppler signal was obtained f r o m the surface of the white paper. from zero to about 6 mm. The laser light was passed to the flow channel through the tissue. T o obtain the Doppler signal the blood was perfused constantly from the reservoir to the collecting device via the flow channel. The intensity of the Doppler signal was measured at various tissue thicknesses.
Signal jluctuation Signal fluctuation, which affects flow measurement, was investigated using a rotating turntable as shown in figure 5. A sheet of white paper was placed on the turntable, and it was rotated at a constant number of revolutions. The probe was placed at a distance of approximately 10 mm from the surface of the paper. The Doppler signal patterns were sequentially recorded by each LDF system.
Accuracy of blood jlow measurement Since it is generally difficult to measure the absolute blood flow velocity in living tissue, it was evaluated using glass capillary tubes as shown in figure 6. The experimental system consisted of a distribution reservoir A, a tube B, a measuring section C, an electromagnetic flowmeter EMF (Nihonkoden Co. Ltd, Japan) and a collecting device D. The inlet of the measuring section was connected to the reservoir and its outlet was 238
connected to the collecting device. The inner diameter of the tube at the EMF was 8.0 mm. The measuring section (figure 6b) consisted of 89 parallel glass capillary tubes (inner diameter 0.2 mm). This was similar to the mechanical fluid model reported by Nilsson et al. [3], which consisted of 37 symmetrically-arranged tubes (inner diameter 0.3 mm) and a 0.2-mm thick slot in a radial direction toward the periphery between the edge of the tubes and the optical sensor. In the present system a wider measuring section was needed to receive the illumination of the light in the Diodopp system, in which the distance between the laser diode and photodetector is relatively large (L=2*73 mm). Consequently, it was necessary to use 89 tubes. Thc distance between the end of the capillaries and the bottom of the tube B, which was the flow channel for flow from the capillaries, was 2.5 mm. The haematocrit of the perfused blood was 12%. Each LDF probe was successively attached to the outer side of the transparent glass plate at the bottom of the tube B (figure 6b). The blood flow was measured at each flow volume, as the blood was perfused at a constant rate.
Results Evaluation of the sampling depth As the sampling depth in blood is affected by the power of the emitted light, the power of the light of each
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements
Table 1. Emitted laser power of each LDF system
Table 2. Maximum detectable depth in blood usins each LDF system in the cases of haematocrit 5.0 and 36.5% (whole blood)
Emitted laser power (mW)
LDF system
1.98 1.99 - 2.02 1.99
Periflux ALF-2 1 Diodopp
LDF system
Haematocrit 5.0°/0
Haematocrit 36.5%
2.0
02
4.0 3.0
2.6 2.0
Periflux ALF-2 1 Diodopp
Downloaded by [RMIT University Library] at 05:34 22 May 2016
system was previously measured with an optical power meter (Model 44XLA, Photodyne Inc., USA). The power of all three systems was confirmed to be nearly equal (= 2 mW), but the stability of the ALF-21 was poor compared to that of the other systems, as shown in table 1.
in which a He-Ne laser (632.8 nm) is used. Compared to the Periflux, the ALF-21 was 2.0 times more sensitive with the lower haematocrit blood and the Diodopp was 1.5 times more sensitive. With the higher haematocrit blood, the value for the ALF-21 was one order of magnitude greater than the Periflux value.
The Doppler signal gradually disappeared as the depth of stored blood was increased. The relationship of the signal intensity to the depth was exponential, as shown in figure 7. The maximum detectable depth is shown by the arrow in each figure, and is shown in table 2 for haematocrits 5.0 and 36.5%. I n both cases the maximum depth was obtained with the ALF-21, followed by the Diodopp system; in both systems a longwavelength (780 nm) semiconductor laser diode is used. The minimum depth was obtained with the Periflux,
With regard to tissue, the relationship of the signal intensity to the depth is shown in figure 8. The maximum depth (dmaX)reached by each system is given in table 3, and indicated by the arrows in figure 8. The maximum depth in the case of blood was obtained with the ALF-21 followed by the Diodopp and Periflux systems, respectively. Compared with the Periflux, the
Reservoir (A)
(a)
7
I
0 Blood
f E
E
0
ow Chamber (6) easuring Section (C)
(0
I
ass Capillary Tubes)
' II
Collecting Device (D)
C30mm-4
Figure 6. (a) Schematic conjguration of the experimental arrangement for determining the accuracy of blood flow measurement. Blood was perfused through the measuring section (C) which was composed of 89 capillary tubes. The laser beam was emitted at the edge of the assembled capillary tubes to obtain the Doppler signal. (b) Enlarged illustration of the capillary tube measuring section.
05-
10
I --.
g
..__
20
30
40
Thickness of Blood Layer (rnrn)
(a)
Thickness of Blood Layer (rnrn)
(b)
Thickness of Blood Layer (rnrn) (C)
Figure 7. Relationship between the detected Doppler signal intensity and the thickness of stored blood (d) in the case of blood haematocrit 5.0%. The signal intensity is given in arbitrary units (A U).The arrows indicate the maximum detectable thickness d,,,. (a) Periyux, (a) ALF-21, (c) Diodopp. 239
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements
1 .o
3
5
0
a a 0
.Q 0
10
20
.. 30
.
0
' e 9 .
0 40
50
60
Thickness of Blood Tissue (mm)
10
20
t 2 .... 40
50
C
10
60
Thickness of Blood Tissue (mm)
(a)
Downloaded by [RMIT University Library] at 05:34 22 May 2016
30
.. *.p. .... 20
30
40
50
60
Thickness of Blood Tissue (mm)
(b)
(C)
Figure 8. Relationship between the detected Doppler signal intensity and the thickness of tissue. The signal intensity is given in arbitrary units ( A U). The arrows indicate the maximum detectable thickness d,,, (a) PeriJux, (b) ALF-PI, (c) Diodopp.
M 1 min
Time
Time
1 mi"
Figure 9. The jluctuation of the Doppler signal when the laser probe is placed perpendicular to a surface of a moving white paper.
ALF-21 was about 2.3 times more sensitive and the Diodopp was about 1.9 times more sensitive. Signal jluctuation The Doppler signal patterns sequentially obtained by each system are shown in figure 9. Small periodic fluctuations of the signal coinciding with the period of the rotational speed of the turntable were observed in each pattern. As this fluctuation was considered to be due to a slight distortion of the paper surface, signals Table 3. Maximum detectable depth in tissue usins each LDF system
Periflux ALF-2 1 Diodopp ~~
240
1.5 3.5 2.8
were sampled for the same temporal period. The coefficient of variation (c.v.) of each fluctuation was 0.53 in the Periflux, 0.81 in the ALF-21 and 0.27 in the Diodopp. The Diodopp system showed the highest stability, and the ALF-21 shows the lowest stability. The lower stability of the ALF-21 may be due to instability of the laser source, as shown in table 1 .
Accuracy of blood Jlow measurement The Doppler signals obtained are shown in figure 10. Arbitrary units were used, and they were standardized by the maximum value in each case. The Periflux and Diodopp systems showed good linearities between the Doppler signal and blood flow volume. However, the linearity between the Doppler signal and blood volume for the ALF-21 was poor. It may be due to the difference of the sampling depth in the flow channel.
K. Mito Charactrristics of lasrr Doppler flowmeters with direring optical arrangements
Downloaded by [RMIT University Library] at 05:34 22 May 2016
Discussion The maximum detectable depth differed for each optical arrangement, although the power of the emitted light from the three systems is nearly equivalent. The maximum value in blood and tissue was obtained with the ALF-2 1, followed by the Diodopp and the Periflux (figure 11). It was confirmed that a LDF system with a long-wavelength semiconductor laser diode (780 nm) can sample at a greater depth than those with an He-Ne laser (632.8 nm). Our results show that differences exist in the optical parameters of both blood and tissue, i.e. an absorbing coefficient, a scattering coefficient and a phase function. Since Van Gemert et al. [ 4 ] suggested that the absorbing and scattering coefficients at a wavelength of 632.8 nm are similar to those at a wavelength of 780 nm, it was considered that the differences were mainly the result of differences in the phase function at short (632.8 nm) and long wavelengths (780 nm).
for blood. As the tissue structure in the sampling volume is probably unchanged, it is assumed that the sampling depth in the practical tissue microcirculation case is affected by the changeable haematocrit in several disease conditions respectively. The variation in the depth obtained was the largest in the Periflux, followed by ALF-21, and the Diodopp, as shown in Figure 11. It is, therefore, suggested that the Diodopp system should be used as the change of haematocrit has the smallest effect in practical blood flow measurement. Jentink et al. [5] reported that the sampling volume of LDF can be extended to deep layers by increasing the distance (L) between the laser beam and detector using the Monte-Carlo simulation. This distance is 0-82 mm in the Periflux system, 0.66 mm in the ALF-21 and 2.73 mm in the Diodopp. However, the maximum detectable depth was obtained with the ALF-21 with the minimum distance L and not with the Diodopp. Therefore, the depth may be affected not only by the distance but also by the numerical aperture.
Figure 11 suggests clearly that the sampling depth in the practical tissue microcirculation case will differ if the optical arrangements of the LDF systems differ. Furthermore, the depth of the measurement of each system will vary with haematocrit. In this study the depth obtained in tissue was within the range of that
With regard to signal fluctuation, the ALF-21 and Periflux systems were not stable. Periodic noise was observed in the signal of the Periflux system. This may be the laser cavity warming-up effect, as reported by Obeid et al. [ 6 ] . The ALF-21 shows instability which
. ..
s
f
0
50
1.0
4
*
...I
100
Blood Flow Volume (EMF) (ndlmin)
Blood Flow Volume (EMF) (nllmin)
Blood Flow Volume (EMF) (nl/min)
(a)
(b)
(C)
Figure 10. Relationship belmeen the Doppler shiJ signal and [he known blood flow volume: (a) PeriJux, (b) ALF-21, (c) Diodopp.
5n
rnrn 4.0
-
40
d Q
0 Tissue
3.0-
3.0
c 0 0 c Q
n
5E
2.6
2.0 -
2.0
.X
i?-
1.0-
0.2
-I
0.’
Periflux
ALF- 21
2.0
Diodopp
Figure 11. The maximum delectable depth in blood and tissue. 24 1
K. Mito Characteristics of laser Doppler flowmeters with differing optical arrangements may be due to fluctuation of the power of the emitted laser light, as shown in table 1.
Downloaded by [RMIT University Library] at 05:34 22 May 2016
As for the accuracy of blood flow measurements, the Diodopp and Periflux systems showed good linearities between Doppler signals and blood flow volumes, but the ALF-21 did not. This may be due to difference in the sampling depth: in the ALF-21, many blood flow measurements, including the flow in the capillary tubes, were obtained in sprayed blood between the edge of the tubes and the inside glass plate, as shown in figure 3, with its large sampling volume.
Conclusion In this study, sampling depth, stability and accuracy clearly differ with different optical arrangements. Although the data obtained in this study are only relative, they provide important information regarding LDF systems with different optical arrangements. In measuring blood flow in the practical tissue microcirculation case, a moderate sampling depth, smaller variance with the change in haematocrit, and stability are necessary for accurate measurement. For this purpose a system which can be directly applied to blood flow using a near-infrared laser light without optical fibres, such as the Diodopp system, seems most promising.
242
Acknowledgement
I express my thanks to M r Osamu Yasuda, General Manager (Nikko Keisoku System Ltd) and Mr Nobuhiko Ueda, Ms Tomoko Fujita (Nikko Keisoku System Ltd), Fumika Nakamura, Shinichi Miyake and Katsuya Tsuda (Kawasaki College of Allied Health Professions) for their cooperation in the experiment. References 1. OBEID,A. N., BOGCETT, D., DOUGHERTY, G., BARNETT, N. J. and ROLGE, P. (1988) Depth discrimination in laser
Doppler skin blood flow measurement using different lasers. Medical and Biological Engineering and Computing, 26, 415-419. 2. OBEID,A. N., DOUGHERTY, G. and PETTINCER, S. (1990) In vivo comparison of a twin wavelength laser Doppler flowmeter using He-Ne and laser diode sources. Journal of Medical Engineering and Technology, 14, 102-1 10. 3. NILSSON, G. E., TENLAND, T. and OBERG,P. (1980) Evaluation of a laser Doppler flowmeter for measurement of tissue blood flow. I E E E Transactions on Biomedical Engineering, BME-27,597-604. 4. VANGEMERT, M. J. C., JACQUES, S. L., STERENBORG, H. J. C. M. and STAR,W. M. (1989) Skin optics. IEEE Transactions on Biomedical Engineering, 56, 1146-1 154. 5. JENTINK, F. F., DE MUL, M., HERMSEN, R. G. A. M., GRAAFF, R. and GREVE, J. (1990) Monte Carlo simulations of laser Doppler blood flow measurements in tissue. Applied Optics, 29, 2371-2381. 6 . OBEID,A. N., BARNETT,N. J., DOUGHERTY, G. and WARD,G. (1990) A critical review of laser Doppler flowmetry. Journal of Medical Engineering and Technbiogy, 14, 178-181.
