Abstract

A system for full-field laser Doppler blood flow imaging has been developed and tested on biomedical samples. The new imaging system allows 2D flow maps or monitoring flux signals to be obtained from a plurality of measured points simultaneously by using a 2D array of photodetectors. The detection part of the system is based on an intelligent CMOS camera with a built-in digital signal processor and memory. The imaging time of the system is as much as to 4 times faster than for the conventional scanning laser Doppler imager. The performance of the system was evaluated in vitro and in vivo. The first measurement results with this new system on human skin are reported.

©2005 Optical Society of America

1. Introduction

Laser Doppler perfusion imaging (LDPI) is an interferometric technique successfully used for visualization of two-dimensional (2D) microvascular flow maps in a number of clinical investigations, including peripheral vascular disease, skin irritants, diabetic conditions, burns, and organ transplants. The technical principle is based on the Doppler effect, where light scattered by moving particles, e.g., blood cells, receives a slight frequency shift, which can be measured by heterodyne detection [1]. A 2D flow map is obtained by means of successive measurements from a plurality of predetermined points. Classical LDPI systems use mechanical scanning of the area of interest with a narrow-collimated or focused laser beam [2, 3]. However this scanning approach is time consuming and suffers from artifacts caused by the mechanical steering of the irradiating laser beam. In the current commercial LDPI systems these artifacts are circumvented at the expense of imaging time.

Fujii et al. [4] and Briers et al. [5] proposed alternative full-field flow imaging techniques that use speckle contrast analysis. The advantage of this approach is fast image acquisition, which is achieved at the expense of spatial and temporal resolution. However, the technique can hardly be exploited for objective flow measurements when either the concentration or the speed of moving particles is not known in advance. Concentration and speed influence the system response and contribute to contrast decay. Both parameters are important for medical diagnosis but cannot be evaluated separately with a laser speckle imaging technique.

The laser Doppler technique gives access to more objective information [6]; i.e., concentration as well as velocity can be measured separately. Recently Serov et al. [7] suggested a new modality for laser Doppler imaging: a true-random-addressing CMOS image sensor was used to detect the Doppler signal from a plurality of points on the sample. For this approach the area of interest is illuminated with a divergent laser beam. Mechanical scanning is replaced with a photoelectrical scan by a 2D array of photodetectors, resulting in faster imaging than with the mechanical scanning approach.

In this paper we further improve the concept of parallel LDPI. A new laser Doppler imager based on a new generation of true-random-addressing CMOS image sensor has been developed. Our approach takes the advantage of the CMOS image sensor pixel architecture coupled with a digital signal processor (DSP) to improve the flow map refresh rate. We optimized the system performance in terms of acquisition speed and computation time. For illumination we used a laser light coupled into a multimode optical fiber, which results in uniform illumination of the target.

We also investigated the sensitivity and the linearity of the imaging system response to changes in concentration and velocity of moving particles and performed several in vivo experiments measuring perfusion on the human skin as blood flow changes as a result of occlusion, deep breath, and cold-water immersion.

2. Experimental Setup

Figure 1 shows a schematic of the setup for parallel laser Doppler flow imaging. For the area illumination we used a 2 mm core diameter multimode optical fiber made of acrylic polymer. Light from a diode-pumped solid-state (DPSS) laser (671 nm, 50 mW output power) was coupled into the fiber. The light from the distal end of the fiber was projected onto the sample with a magnification lens, achieving homogeneous illumination of the area of interest. The size of the illuminating spot, typically several centimeters in diameter, could be varied with the magnifications lens.

For observing the sample and detecting the Doppler signal an intelligent CMOS camera was used (iMVS-155, Fastcom Technology SA, Switzerland). It was placed at a fixed distance of approximately 20 cm above the measured sample. This camera is based on the CMOS image sensor Fuga1000 from FillFactory NV (Belgium). It comprises a matrix of 1024×1024 pixels with a pixel size of 8×8 µm2 and a fill factor of 70%. The analog-to-digital converter (ADC) has a sampling rate of 10 MHz, converting the analog signal to an 8 bit digital signal. The electronic architecture of the CMOS sensor allows a window of interest (WOI) to be acquired at higher frame rates—several kilohertz, depending on the size of the WOI. The fast pixel sampling allows the detection of the intensity variations induced by the Doppler-shifted light, which are typically in the range 0–20 kHz. The signal variations measured at each sampled pixel are stored in the camera memory (4.5 Mbytes in total) and thereafter processed with a 32 bit floating-point digital processor (ADSP 21061 SHARC 40 MHz, Analog Device).

 figure: Fig. 1.

Fig. 1. Experimental setup for full-field laser Doppler flow imaging.

Download Full Size | PPT Slide | PDF

The signal processing includes calculation of the zero-(M 0) and first-moment (M 1) of the spectral power density S(ν) of the intensity fluctuations I(t) over the WOI. The average concentration, 〈C〉, of moving particles in the sampling volume can be derived from the zero-moment of the power spectrum. The first moment, which in laser Doppler flowmetry is called flux or perfusion, is proportional to the root-mean-square speed of moving particles, V rms, times their average concentration [8]:

Perfusion=CVrmsM1=0νS(ν)dν,
Concentration=CM0=0S(ν)dν,
SpeedM1M0,
S(ν)=0I(t)exp(i2πνt)dt2.

Here ν is the frequency of the intensity fluctuations induced by the Doppler shift. We calculate the power spectrum by using a fast Fourier transform (FFT) algorithm applied to the recorded signal variations at each sampled pixel of the WOI. Finally, the perfusion and the concentration maps are calculated and shown on the display.

The strength of the photodetector signal generated by the dynamic speckle pattern depends on the average speckle size and the average brightness of the detected speckle pattern. In practice, the optimal detection conditions for the full-field laser Doppler approach differ from the optimal detection conditions for laser speckle contrast analysis. The sampling speed in the laser Doppler approach is of the order of several microseconds, whereas for the speckle contrast analysis the integration time is several milliseconds. Thus the laser Doppler approach is more demanding on the brightness of the detected light. To increase the signal-to-noise ratio (SNR), a low-f-number TV lens has to be used to collect as much scattered light as possible. A high-f-number TV lens would increase the average speckle size on the detector, and, consequently, the amplitude of the signal oscillations would be stronger, but on the other hand it would require a substantial increase of the illuminating laser power, which is limited for safety. For our imager we used an 8.5 mm focal distance TV lens with an f-number of 1.3. The estimated average number of speckles per pixel was approximately 40.

Since our system is not sensitive to room ambient light, no color filter was used. Thus black and white photographic images of the object of interest are obtained with the same CMOS camera. This image is useful for determining the anatomical boundaries associated with the perfusion regions presented in the blood flow maps.

In the monitoring mode the imager allows instantaneous variations of the perfusion signal to be measured at one point on the object. In this mode a perfusion trace is displayed on the screen in real time.

3. Experiments and Results

3.1. Flux Response Test: Velocity and Concentration

For the flux response performance test a special artificial sample (phantom) was made of white Teflon with optical scattering properties close to the optical properties of human skin. This Teflon block contains a circular channel 1 mm in diameter, placed at a depth of 1 mm below the illuminated surface. The flow through the channel was varied in a controlled manner with a liquid pump. The pump allowed the average velocity of the fluid in the channel to be varied over the range from 0 to 3 mm/s. As a liquid scattering sample we used an Intralipid-20% solution at various concentrations. The original Intralipid-20% suspension was diluted to concentrations of 0.5%, 2.5%, and 5%. Thus the absolute concentrations of Intralipid in the solutions were 0.1%, 0.5%, and 1%, respectively. The sample was illuminated with a 50 mW laser beam over an area 20 mm in diameter. The Doppler signal was recorded with a 40 kHz sampling frequency from a single pixel and processed.

 figure: Fig. 2.

Fig. 2. Flux response of the CMOS imager in respect to velocity measured for different Intralipid concentrations.

Download Full Size | PPT Slide | PDF

In Fig. 2 the dependence of the CMOS imager flux response to the velocity is shown for three different Intralipid concentrations. The plot shows a linear flux response of the imager for velocities up to 3 mm/s, which was the maximum velocity provided by the pump. The measured flux response to concentration also demonstrates an increase in the flux value. However Fig. 2 shows that the measured flux response is not linearly proportional to the concentration. We explain this observation as a photon multiple scattering effect: the first moment is linearly proportional to the concentration of moving particles, provided that this concentration is small [8], which is the case for many biological tissues. In the case of a high concentration of moving particles, multiple scattering occurs and leads to a nonlinear flux response of the system [9]. The scattering phantom applied here for the calibration is characterized by multiple scattering of the light. For the Intralipid concentrations used, the scattering coefficients at a wavelength of 671 nm are µs=0.4, 2, 4 mm-1 [10]. Considering the scattering geometry of the test phantom, multiple scattering on the Intralipid particles is expected for a 1 mm thick channel placed 1 mm below the illuminated surface.

One can note that the flux response of the imager is not zero for zero flow velocity; this is due to the Brownian motion of the scattering particles in the Intralipid solution.

3.2. In Vivo Performance: Real Time Monitoring of Blood Flow

In order to monitor the perfusion changes in real time, the Doppler signal was measured for one pixel sampled at 40 kHz on a fingernail illuminated by a 50 mW laser beam of 40 mm in diameter. Figure 3 shows the measured perfusion time traces obtained from a single pixel, measured and displayed in real time. These single-point measurements give a high temporal resolution (10 Hz chart data flow rate is typical), allowing rapid blood flow changes to be recorded.

 figure: Fig. 3.

Fig. 3. Time traces of the perfusion signal obtained with the full-field laser Doppler imaging system. The decays in the perfusion signal are due to (left) the indicated occlusion of the upper arm and (right) the indicated deep breath, a so-called Valsava maneuver.

Download Full Size | PPT Slide | PDF

The perfusion time trace shown in Fig. 3 (left) demonstrates the perfusion response caused by an artery occlusion of the upper arm. Since the inflation of the occlusion cuff takes a certain time, the perfusion decreases slowly until the occlusion attains biological zero. After releasing the occlusion the perfusion signal recovers the initial value. We also measured the perfusion response when the subject performed a deep breath test. The measured perfusion time trace is shown in Fig. 3 (right). Directly after the deep breath the perfusion decreases and returns to the initial value after about 10 s. This is the so-called Valsava maneuver that leads to an increase of intrathoracic pressure, causing a temporary decrease of venous return of blood to the heart. Consequently the microcirculation is stopped for a short time.

3.3. In Vivo Performance: 2D Perfusion Imaging

To demonstrate the imaging performance of our apparatus, we acquired the perfusion maps of the finger before, during, and after occlusion of the arm artery. An inflatable cuff, used for blood pressure measurements, is pumped to a pressure of about 180 mmHg to stop the blood flow. The finger was illuminated with a 40 mm diameter laser beam of 50 mW output optical power. An area corresponding to 256×256 pixels was photoelectrically scanned. For the fast sampling rate, the 256×256 pixel window was subdivided into a number of smaller subwindows of 64×8 pixel size, which were sampled 256 times one after the other. Thus the sampling rate for each pixel was 16.8 kHz, which corresponds to the subwindow (64×8) frame rate. To calculate the power spectra of the measured intensity fluctuations, a FFT was applied to the recorded signals. Finally, the first moment of the signal fluctuations at each of 256×256 pixels was calculated according to Eq. (1), and the perfusion map was displayed on the monitor. In total, acquisition, signal processing, and data display took approximately 90 s to yield one perfusion image of 256×256 pixel size. Typical examples of perfusion images are shown in Figs. 4, 5, and 6.

 figure: Fig. 4.

Fig. 4. 128×256 pixel perfusion images of the finger obtained before, during, and after occlusion of the upper arm. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.

Download Full Size | PPT Slide | PDF

In Fig. 4 the measured perfusion maps are shown. As expected, there is a decrease of the perfusion signal during the occlusion. After the occlusion is released, the local perfusion rises above the initial value; this effect is known as reactive hyperemia.

The highly perfused area in Fig. 5 contains a small wound, indicated by the arrow on the black and white image of the finger. The perfusion signal decreases during the occlusion. After the occlusion has been released, the blood flow returns to the initial state with a certain delay time.

Finally, we demonstrate the response of the finger perfusion submitted to a cooling cycle. For 2 min the finger was immersed in ice water. Then we measured the perfusion maps immediately after, 3 min after, and 10 min after the cold bath. In Fig. 6 the obtained perfusion maps are shown. The low temperature has the effect of decreasing the blood flow. After 3 min an increase in perfusion compared with the initial state is observed. After 10 min the perfusion attains its initial state.

One can note that the perfusion images shown in Figs. 46 contain broad dark-blue regions at the edges of the finger. In those regions the exact perfusion value is unknown owing to the low SNR of the measurement. The measured perfusion signal is virtually zero in those regions; however, the regions are not necessarily poorly perfused. Here the low SNR is caused by a shadowing effect (edge effect), since a volumetric object was illuminated at an increasing angle relative to the detection direction; see Fig. 1. This leads to a lower backscattered light intensity measured from the shadowed regions and, consequently, to the lower SNR of the measurement.

Next, blue indicates the perfusion level corresponding to an occlusion level, so-called biological zero. The discrepancy between the biological zero and the real zero output level of the instrument is attributed to the fact that, though the blood perfusion is arrested by the inflated cuff, the blood cells in the peripheral vessels are still moving randomly (Brownian motion), giving rise to minor Doppler components recorded by the instrument.

 figure: Fig. 5.

Fig. 5. 64×64 pixel perfusion images of the finger obtained before, during, and after occlusion of the upper arm. The arrow indicates a small wound on the finger where the perfusion is altered by healing. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.

Download Full Size | PPT Slide | PDF

 figure: Fig. 6.

Fig. 6. 256×256 pixel perfusion images directly after, 3 min after, and 10 min after the immersion of the index finger in the ice water. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.

Download Full Size | PPT Slide | PDF

We estimated the SNR of our system up to 8 dB for measurements on the finger. For those measurements the root-mean-square (RMS) value of the noise signal was measured on a statically scattering object, while the RMS value of the signal was measured directly on the finger under normal conditions.

4. Discussion

The results presented in this paper demonstrate a reliable performance of a new parallel laser Doppler imager applied to a human finger. With 0.16 mW/mm2 illuminating intensity we achieved a SNR up to 8 dB. Here the finger was chosen because the finger perfusion is known to be fairly high, compared with the skin of the forearm. The illuminating intensity of our system was 10 times lower than typically used in conventional scanning imagers. Obviously, increasing the illuminating laser power will proportionally increase the SNR of the system. Also we find that, even if the shadowing effect can cause some artifacts, those artifacts can be reduced either by illuminating the object from several directions or by an increase in illumination power.

We demonstrated the system performance for 2D perfusion imaging as well as for real-time monitoring of the perfusion signal at one point. The system shows a linear response to the velocity, tested up to 3 mm/s with an illumination power of 50 mW over an area of 20 mm in diameter. Actually the upper frequency cutoff is limited not only by the sampling rate of the sensor but also by the pixel response time. This pixel response time has its origin in the nonintegrating nature of the sensor; the stronger the light, the shorter the pixel response time [11]. Thus the bandwidth also depends of the intensity of the detected backscattered light. Here the pixel response time plays the role of an analog antialiasing filter, if the pixel response time is longer than the sampling time. The lower frequency cutoff is limited by the amount of memory available for storage of the acquired subframes. For example, for the measurements presented in this paper we had a bandwidth from 40 to 20 kHz for real-time monitoring of the flow and from 66 to 8.4 kHz for the perfusion imaging.

The refresh rate of the perfusion images is about 90 s for the 256×256 pixel window. This time includes acquisition, signal processing, and transferring the data to the display. Most of the time is taken by the FFT calculations performed with the integrated DSP. For comparison, the specified scan speed of a commercial laser Doppler imaging system MoorLDI (Moor Instruments Ltd, UK) is typically 20 s for a 64×64 pixel resolution and 5 min for a 256×256 pixel resolution. This is 3–4 times slower than our imaging system. However, the scanning imager can measure areas of up to several decimeters (50 cm×50 cm) in size, while with the area illumination approach the imaged area is limited to several centimeters owing to the lower illuminating intensity.

The logarithmic intensity response, fixed patter noise (FPN), and 8 bit ADC resolution reduce the SNR of our system. The logarithmic response results in a very high intrascene dynamic range (120 dB) but also in a high-frequency noise and FPN caused by the dark current variations from pixel to pixel. If the logarithmic intensity response is essentially a property of the true-random-addressing pixel architecture, the FPN performance and ADC resolution can obviously be improved in next sensor generations.

In summary, our imager has several specific advantages versus the alternative imaging systems:

• 3–4 times faster imaging time than the classical scanning imagers;

• Direct measurements of the flow velocity, which is not the case for the laser-speckle flow measurement techniques;

• Independence of the speed and concentration instrument response, which is also not the case for the laser-speckle imaging;

• Higher temporal and spectral resolution (around 50 Hz) and larger bandwidth (up to 20 kHz) comparing with the laser-speckle flow measurement techniques.

The new imaging approach is a good compromise between the imaging speed and the measurement accuracy. However the present performance of the imager does not allow imaging of the areas larger than approximately 50 mm×50 mm. Thus at the state of the art the imager could be used for some specific applications that do not require measurements over large areas, e.g., for dermatology, to differentiate between various types of skin tumors and recording of the blood flow caused by the allergic reactions; for development of skin care products, to visualize possible adverse effects of substances applied to the skin; and for wound healing, to see the response to different therapies for ulcer healing.

5. Outlook

The CMOS image sensor market is developing very fast. New sensors with improved photoelectrical parameters are already announced. We believe that with the next generation CMOS image sensors even better quality 2D flow maps will be obtained. Also, the refresh rate of the flow images can be increased by using a faster DSP integrated into the camera or even by accomplishing the signal processing in a host PC having a fast data communication with the camera.

Definitely the main advantage of the parallel imaging approach is faster operation speed, which could allow the observation of high-resolution perfusion images with a refresh rate of less than a second.

References and Links

1. A. P. Shepherd and P. Å. Öberg, Laser-Doppler Blood Flowmetry (Kluwer Academic, Boston, 1990).

2. K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993). [CrossRef]   [PubMed]  

3. T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991). [CrossRef]   [PubMed]  

4. H. Fujii, K. Nohira, Y. Yamamoto, H. Ikawa, and T. Ohura, “Evaluation of blood flow by laser speckle image sensing. Part 1,” Appl. Opt. 26, 5321–5325 (1987). [CrossRef]   [PubMed]  

5. J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999). [CrossRef]  

6. J. D. Briers, “Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging,” Physiol. Meas. 22, R35–R66 (2001). [CrossRef]  

7. A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002). [CrossRef]  

8. R. Bonner and R. Nossal, “Model for laser Doppler measurements of blood flow in tissue,” Appl. Opt. 20, 2097–2107 (1981). [CrossRef]   [PubMed]  

9. A. Serov, W. Steenbergen, and F. F. M. de Mul, “Prediction of the photodetector signal generated by Doppler-induced speckle fluctuations: theory and some validations,” J. Opt. Soc. Am. A 18, 622–630 (2001). [CrossRef]  

10. S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992). [CrossRef]  

11. E. R. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices 44, 1698–1698 (1997). [CrossRef]  

References

  • View by:

  1. A. P. Shepherd and P. Å. Öberg, Laser-Doppler Blood Flowmetry (Kluwer Academic, Boston, 1990).
  2. K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
    [Crossref] [PubMed]
  3. T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991).
    [Crossref] [PubMed]
  4. H. Fujii, K. Nohira, Y. Yamamoto, H. Ikawa, and T. Ohura, “Evaluation of blood flow by laser speckle image sensing. Part 1,” Appl. Opt. 26, 5321–5325 (1987).
    [Crossref] [PubMed]
  5. J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
    [Crossref]
  6. J. D. Briers, “Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging,” Physiol. Meas. 22, R35–R66 (2001).
    [Crossref]
  7. A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
    [Crossref]
  8. R. Bonner and R. Nossal, “Model for laser Doppler measurements of blood flow in tissue,” Appl. Opt. 20, 2097–2107 (1981).
    [Crossref] [PubMed]
  9. A. Serov, W. Steenbergen, and F. F. M. de Mul, “Prediction of the photodetector signal generated by Doppler-induced speckle fluctuations: theory and some validations,” J. Opt. Soc. Am. A 18, 622–630 (2001).
    [Crossref]
  10. S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
    [Crossref]
  11. E. R. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices 44, 1698–1698 (1997).
    [Crossref]

2002 (1)

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
[Crossref]

2001 (2)

1999 (1)

J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
[Crossref]

1997 (1)

E. R. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices 44, 1698–1698 (1997).
[Crossref]

1993 (1)

K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
[Crossref] [PubMed]

1992 (1)

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

1991 (1)

T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991).
[Crossref] [PubMed]

1987 (1)

1981 (1)

Bonner, R.

Briers, J. D.

J. D. Briers, “Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging,” Physiol. Meas. 22, R35–R66 (2001).
[Crossref]

J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
[Crossref]

Byrne, P. O.

T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991).
[Crossref] [PubMed]

de Mul, F. F. M.

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
[Crossref]

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Prediction of the photodetector signal generated by Doppler-induced speckle fluctuations: theory and some validations,” J. Opt. Soc. Am. A 18, 622–630 (2001).
[Crossref]

Essex, T. J. H.

T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991).
[Crossref] [PubMed]

Flock, S. T.

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Fossum, E. R.

E. R. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices 44, 1698–1698 (1997).
[Crossref]

Fujii, H.

He, X. W.

J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
[Crossref]

Ikawa, H.

Jacques, S. L.

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Jakobsson, A.

K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
[Crossref] [PubMed]

Nilsson, G .E.

K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
[Crossref] [PubMed]

Nohira, K.

Nossal, R.

Öberg, P. Å.

A. P. Shepherd and P. Å. Öberg, Laser-Doppler Blood Flowmetry (Kluwer Academic, Boston, 1990).

Ohura, T.

Richards, G.

J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
[Crossref]

Serov, A.

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
[Crossref]

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Prediction of the photodetector signal generated by Doppler-induced speckle fluctuations: theory and some validations,” J. Opt. Soc. Am. A 18, 622–630 (2001).
[Crossref]

Shepherd, A. P.

A. P. Shepherd and P. Å. Öberg, Laser-Doppler Blood Flowmetry (Kluwer Academic, Boston, 1990).

Star, W. M.

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Steenbergen, W.

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
[Crossref]

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Prediction of the photodetector signal generated by Doppler-induced speckle fluctuations: theory and some validations,” J. Opt. Soc. Am. A 18, 622–630 (2001).
[Crossref]

van Gemert, M. J. C.

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Wårdell, K.

K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
[Crossref] [PubMed]

Wilson, B. C.

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Yamamoto, Y.

Appl. Opt. (2)

IEEE Trans. Biomed. Eng. (1)

K. Wårdell, A. Jakobsson, and G .E. Nilsson, “Laser Doppler perfusion imaging by dynamic light scattering,” IEEE Trans. Biomed. Eng. 40, 309–316 (1993).
[Crossref] [PubMed]

IEEE Trans. Electron Devices (1)

E. R. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices 44, 1698–1698 (1997).
[Crossref]

J. Biomed. Eng. (1)

T. J. H. Essex and P. O. Byrne, “A laser Doppler scanner for imaging blood flow in skin,” J. Biomed. Eng. 13, 189–193 (1991).
[Crossref] [PubMed]

J. Biomed. Opt. (1)

J. D. Briers, G. Richards, and X. W. He, “Capillary blood flow monitoring using laser speckle contrast analysis (LASCA),” J. Biomed. Opt. 4, 164–175 (1999).
[Crossref]

J. Opt. Soc. Am. A (1)

Lasers Surgery Med. (1)

S. T. Flock, S. L. Jacques, B. C. Wilson, W. M. Star, and M. J. C. van Gemert, “Optical properties of intralipid: a phantom medium for light propagation studies,” Lasers Surgery Med. 12, 510–519 (1992).
[Crossref]

Opt. Lett. (1)

A. Serov, W. Steenbergen, and F. F. M. de Mul, “Laser Doppler perfusion imaging with a complimentary metal oxide semiconductor image sensor,” Opt. Lett. 25, 300–302 (2002).
[Crossref]

Physiol. Meas. (1)

J. D. Briers, “Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging,” Physiol. Meas. 22, R35–R66 (2001).
[Crossref]

Other (1)

A. P. Shepherd and P. Å. Öberg, Laser-Doppler Blood Flowmetry (Kluwer Academic, Boston, 1990).

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1.
Fig. 1. Experimental setup for full-field laser Doppler flow imaging.
Fig. 2.
Fig. 2. Flux response of the CMOS imager in respect to velocity measured for different Intralipid concentrations.
Fig. 3.
Fig. 3. Time traces of the perfusion signal obtained with the full-field laser Doppler imaging system. The decays in the perfusion signal are due to (left) the indicated occlusion of the upper arm and (right) the indicated deep breath, a so-called Valsava maneuver.
Fig. 4.
Fig. 4. 128×256 pixel perfusion images of the finger obtained before, during, and after occlusion of the upper arm. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.
Fig. 5.
Fig. 5. 64×64 pixel perfusion images of the finger obtained before, during, and after occlusion of the upper arm. The arrow indicates a small wound on the finger where the perfusion is altered by healing. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.
Fig. 6.
Fig. 6. 256×256 pixel perfusion images directly after, 3 min after, and 10 min after the immersion of the index finger in the ice water. The six-level color scale representing relative low-to-high tissue perfusion is displayed below the images.

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

Perfusion = C V rms M 1 = 0 ν S ( ν ) d ν ,
Concentration = C M 0 = 0 S ( ν ) d ν ,
Speed M 1 M 0 ,
S ( ν ) = 0 I ( t ) exp ( i 2 π ν t ) d t 2 .

Metrics