We report a novel system for time-resolved diffuse remission spectral measurements, based on short light continuum pulses generated in an index-guided crystal fiber, and a spectrometer-equipped streak camera. The system enables spectral recordings of absorption and reduced scattering coefficients of turbid media in the wavelength range 500–1200 nm with a spectral resolution of 5 nm and a temporal resolution of 30 ps. The optical properties are calculated by fitting the solution of the diffusion equation to the time-dispersion curve at each wavelength. Example measurements are presented from an apple, a finger and a pharmaceutical tablet.
© 2004 Optical Society of America
Over the last decade there has been a lot of effort to develop techniques to extract the absorption and sometimes scattering properties of turbid media. Several areas of research need a tool for such measurements - biomedical applications including tissue diagnostics and physiological measurements, the pharmaceutical industry for on-line measurements of active substance concentration in pharmaceutical preparations, and the food industry for non-destructive measurements of the quality of products. Two, generally different approaches, have been employed to do this. The first one is based on measurements from which it is possible to extract the absorption and scattering properties more or less independently. The measurements can be time-resolved, spatially-resolved or be made in the frequency domain with a modulated light source. The other approach uses spectrally-resolved measurements and utilizes multivariate analysis following training on a large set of known samples. The latter method has been developed mainly within NIR spectroscopy, where scattering has been seen as a complication, yielding uncertainties in the evaluation of the recorded data [1,2]. Large training sets and complex calibration models have often been necessary to span the entire region of the important parameters of the samples under investigation . This approach has two major limitations - it needs frequent calibrations and it is not very robust.
In many respects it is more appealing to use the model-based approach to obtain the absorption and/or scattering properties [4–6]. Frequently time-resolved techniques have been employed (see, for example [7–12]), but also frequency domain [13–19] and spatially resolved [20–22] techniques have been utilized. Such measurements are most often performed at multiple wavelengths in order to enable the desired information to be obtained. This can be achieved with diode lasers at fixed wavelengths or by scanning a tunable laser over the wavelength region of interest [23,24]. Parallel detection of all wavelengths of interest is possible with a short-pulsed broad light source. Such a suitable source is continuum generation employing non-linear interaction in a special optical medium as a result of high peak-power illumination [25,26]. A system based on this technology has been developed in our laboratory. It relies on focusing a high-power laser beam in a sapphire crystal or a cuvette of water [11,27,28]. The laser system in that work runs at 10 Hz, and some averaging is required to achieve a sufficient signal-to-noise ratio to extract the absorption properties with reasonable accuracy. This results in a substantial acquisition time and the time-jitter between the pulses also reduces the time resolution of the system.
Light continuum generation has been utilized for many spectroscopy applications. Recently, non-linear effects in microstructured fibers, designed to have a very low dispersion and thus retain a high pulse peak power throughout the full length of the fiber, have been employed for this purpose [29,30]. Since the non-linear efficiency is very high for these fibers, it is sufficient to use it the with moderate peak powers, obtained by mode-locked lasers. Thus a relatively compact new light source is available for spectroscopy of turbid media .
The objectives of this study were twofold: first to demonstrate the function of a complete time-resolved spectroscopy system based on continuum generation in an index-guiding crystal fiber in the entire wavelength range 500–1200 nm, covering most of the wavelengths of interest for the applications mentioned above. Next, we show measurements conducted on three types of samples to demonstrate the capability of the system.
2. Material and methods
2.1 System description
The arrangement of the system is depicted in Fig. 1. The Ar-ion laser pumped mode-locked Ti:Sapphire laser produced pulses shorter than 100 fs at a repetition rate of 80 MHz. The wavelength of the laser light was centered around 800 nm, and the energy of each pulse was 4 nJ. An optical isolator was used after the laser, to prevent optical reflections that provide unwanted feedback to the laser causing unstable output conditions. A prism compressor was used in the set-up to compensate for the time dispersion caused by the different optical components. The light output from the compressor was focused into a 100 cm long index-guiding crystal fiber (ICF) (Crystal Fiber A/S, Copenhagen, Denmark) using a conventional x40 microscope objective lens with a numeric aperture of 0.65. The ICF had a core diameter of 2 µm and was manufactured to have minimum dispersion at 760 nm. A light continuum was generated by employing non-linear optical effects in the fiber, mainly self-phase modulation and stimulated Raman scattering . A low dispersion of the light inside the fiber combined with small core diameter results in a high peak power of the light in the entire length of the fiber, yielding a high non-linear efficacy resulting in widely spectrally broadened light emission. As a result of this, light pulses with a spectral width spanning from 500 nm to at least 1200 nm were accessible. The light distribution was, however, not perfectly flat throughout the entire wavelength region, but relatively modulated. An advantage of this technique is that it is totally independent of such modulations as long as sufficient light intensity was obtained for all wavelengths of interest, since the optical properties are derived from the time dispersion curves in the sample.
For the sample measurements, either of three geometries was employed as indicated in Fig. 1. Most samples have been measured with a fiber-based probe. The light from the distal end of the ICF is coupled into a 600 µm diameter gradient index fiber that is held in contact with the sample to be measured. Another fiber was used to pick up the remitted light at a certain position of the sample. This was performed either in reflectance or in transmission mode. A third probe geometry, employed mainly for measurements of pharmaceutical tablets, utilized a non-contact beam arrangement. Here, light from the distal end of the ICF was slightly focused onto the tablet surface using a lens. The spot size on the tablet was approximately 2 mm. The tablet was held into place by a circular iris holder, suppressing any light scattered away outside the tablet. Another lens system was used to image the transmitted light onto the entrance slit of the detection system. The detection comprised an imaging spectrometer and a streak camera, yielding spectrally and temporally resolved data in the wavelength region from 500 to 1200 nm. A 25 cm imaging spectrometer (Chromex, Model 250 IS) was equipped with an adjustable entrance slit and three gratings with 30 to 150 grooves/mm. If the entire wavelength range was to be measured, it was necessary to take more than one recording, with different positions of the grating. The spectrally dispersed light at the output of the spectrometer was captured by the streak camera (Hamamatsu, Model C5680). The streak tube utilized an S1 photocathode in order to cover the entire wavelength range of interest. The streak camera operated in synchro-scan mode, allowing all light pulses to be collected. A small portion of the laser light was redirected by a beam splitter onto a photodiode that triggered the streak camera sweep. The system had a total temporal range of 2.1 ns with resolution of 4.5 ps. The instrumental response function was in the range of 30 ps when averaging over 50 s.
2.2 Measurement procedure
In preparation of each measurement session, the source optics needed alignment. This is due to the small diameter of the ICF and a long distance between the pump laser and the source in our arrangement. Misalignment resulted in reduced output power with a less broad spectral profile. Firstly, the optical coupling into the ICF was optimized by adjusting the position of the input end of the fiber with the use of an XYZ translation mount. A routine for this adjustment emerged, where the visually observed intensity of the green light generated in the fiber was used in the adjustments . A better optical coupling resulted in higher light intensity within the fiber and thus higher peak power and better non-linear efficiency. The next step was to adjust the dispersion of the pulse by the prism compressor. This was again adjusted by observing the green light generated in the fiber. After an iterative procedure, a measured spectral profile as illustrated in Fig. 2 was obtained. Small adjustments resulted in significant changes in both intensity and spectral profile of the light continuum.
Prior to each sample measurement, an instrumental response function was recorded. This was either done by connecting two fibers to each end of a thin metal tube indented at the middle to decrease the light measured, or by inserting a pin-hole in the light path. The light intensity was further reduced by inserting absorbing glass filters. The instrumental response function was important in the subsequent analysis to determine the time of the laser pulse in the camera streak without the dispersion caused by the sample and to measure the dispersion of the measured pulse due to the system characteristics.
The next step was to insert the sample to be examined and adjust the recorded signal level. The main adjustments were made by setting the integration time of the CCD camera. The time was normally set between one and five seconds, while typically 50–100 readouts were accumulated in the computer before analysis. This resulted in an acquisition time of approximately one to five minutes. For very high light levels, an optical neutral density filter was employed to reduce the recorded light intensity. Other parameters influencing the recorded signal level was kept constant between measurements, not to alter the characteristics of the detection. The gain of the multi channel plate (MCP) in the streak camera was thus kept fixed, so that the dark current level in the camera system would remain the same for all measured samples. The slit widths of the streak camera and spectrometer were also kept at fixed widths, 50 and 100 µm, respectively.
2.3 Sample measurements
Before the system was utilized for measurements on various samples, its potential to extract optical properties from a homogenous turbid medium was evaluated. For this check, four tissue phantoms as prepared according to Ref.  were employed. The 6.5 cm diameter and 5.5 cm thick epoxy phantoms contained TiO2 particles as scattering centers and toner powder as absorber. The phantoms were measured in a diffuse reflectance geometry using a 1.0 meter long 600 µm core diameter gradient index fiber as a source and collection fiber, respectively. The inter-fiber distance during the recordings was 8 mm. As a gold standard for the determination of expected optical properties, integrating sphere measurements of 1.00 mm thick samples, prepared simultaneously as the phantoms were used .
Following these initial measurements, a number of samples were studied, to illustrate the potential of this system in determining absorption and scattering spectra of turbid samples. Firstly, apples were analysed. A small piece of a fresh Golden Delicious apple obtained from a nearby grocery store was removed, creating a flat skinless surface with a diameter of approximately 30 mm. The apple was measured immediately after the cut with the same diffuse reflection geometry as used for the phantoms.
Next, the tip of the index finger of a volunteer was measured. The finger was measured in transillumination using the same fibers as above for light delivery and collection. The measurement was conducted through the nail. The thickness of the finger was 7 mm. The evaluation was conducted assuming transmission through a infinite slab of thickness 7 mm. Finally, a pharmaceutical tablet (typical immediate release tablet, AstraZeneca R&D Mölndal, Sweden) especially prepared for optical measurements was examined. The tablet was produced in a cylindrical shape with flat end surfaces and a diameter of 13 mm and a thickness of 2 mm. For this measurement, the light continuum from the crystal fiber was slightly focused with a microlens to form a convergent beam with a diameter of 2 mm at the surface of the tablet. The diffuse light transmitted through the tablet was collected and focused onto the entrance slit of the spectrometer using two achromatic lenses.
2.4 Evaluation of absorption and scattering spectra
The recorded data images were evaluated as indicated in Fig. 3. A recorded image contained temporally (x-axis) and spectrally (y-axis) resolved information of the remitted light, accumulated from a sample during typically one minute. The optical properties were then analyzed at each wavelength independently by fitting the experimental data to an analytical solution of the diffusion approximation of the transport equation for a homogeneous semi-infinite medium or an infinite slab . In the evaluation procedure, boundaries are accounted for by employing an extrapolated boundary condition . The best fit was reached iteratively with a Levenberg-Marquardt algorithm, where µs’, µa and an overall amplitude factor are varied in order to minimize a χ 2 merit norm. The temporal shift between the IRF and experimental data is known and is thus not regarded as a free fit parameter. Each iteration involves a convolution between the theoretical time-dispersion curve and the IRF. The fitting range included all points with a number of counts higher than 80% of the peak value on the rising edge of the curve and 1% on the tail . A typical outcome is presented in Fig. 3, and an example of the proceedings of the algorithm is given in Fig. 4.
3. Results and discussion
An estimation of accuracy in the evaluation of optical properties from the recordings of the system is given by correlating with those estimated for four phantoms, produced with different absorption and scattering properties. A correlation plot is presented in Fig. 5, illustrating the agreement between estimated and evaluated optical properties at 916 nm. The estimated values were obtained in agreement with integrating sphere measurements of thin slabs of the phantoms and time-resolved measurements at a specific wavelength using a time-correlated single photon counting system . As can be seen the obtained values agree within approximately 10% with the estimated values for the absorption and within 20% for the scattering.
Spectra of optical properties from an apple fruit are illustrated in Fig. 6a. As can be seen the reduced scattering coefficient decreases almost linearly with increasing wavelength. By fitting the spectrum to the expression for reduced scattering coefficient as a function of wavelength, µs’=aλ-b, where b=0.3, it is obvious from Mie theory that the size of the effective scattering centers in the apple are relatively large. The absorption spectrum is dominated by water absorption peaking around 975 nm.
An example of results from a biomedical research related recording is given Fig. 6b. The spectra are recorded in transmission geometry for an index finger. The slope of the scattering is slightly higher than for the apple, indicating that the effective size of the scattering centers within the finger is smaller. Here, the b-factor is evaluated to be b=0.9. The main absorber in the presented wavelength band is water. The absorption spectrum does not show any lipid content in the finger, which would have been visible close to 915 nm.
A last spectrum illustrates a typical pharmaceutical example. In Fig. 7 an evaluated absorption spectrum of a pharmaceutical tablet especially produced for these measurements in order to obtain a thin tablet and simple measurement geometry, is shown. The absorption spectrum is in good agreement with the active substance in the tablet . The scattering coefficient for this tablet was about 500 cm-1.
The source of the system developed is based on a light continuum generated in a crystal fiber. The crystal fiber technology is rapidly developing and key features of continuum generation in such fibers, such as spectral profile and bandwidth, are quickly improving. The system described in this paper is, however, relatively insensitive to the exact spectral profile of the continuum. As long as the light is sufficiently high at each wavelength of interest, the optical properties can, as seen above, be obtained from a measurement. The detection unit of the system comprises a spectrometer and a streak camera. The spectrometer allows the recording of a wide spectral range, with a relatively high resolution. This is very important for most NIR spectroscopy applications. The streak-camera, on the other hand, provides a very high temporal resolution, enabling measurements of relatively low dispersion objects. As compared to a time-correlated single photon counting system used by several other groups in time-resolved diffuse remission spectroscopy, this system provides a unique combination of relatively short acquisition time in combination with high spectral and temporal resolution.
The authors would like to thank Fabien Chauchard, Cemagref, Montpellier, France, for assistance in the measurements and discussions regarding the fruit samples. We are also grateful to Anders Persson for keeping the performance of the laser at a very high level. The work was financially supported by AstraZeneca R&D Mölndal, Sweden, the Swedish Research Foundation, and the Swedish Research Council.
References and links
1. P. Geladi, D. MacDougall, and H. Martens, “Linearization and scatter correction for near-infrared reflectance spectra of meat,” Appl. Spectrosc. 39, 491–500 (1985). [CrossRef]
2. S. Wold, H. Antii, F. Lindgren, and J. Ohman“Orthogonal signal correction of near-infrared spectra,” Chemom. Intell. Lab. Syst. 44, 175–185 (1998). [CrossRef]
3. V. Centner, J. Verdú-Andrés, B. Walczak, D. Jouan-Rimbaud, F. Despagne, L. Pasti, R. Poppi, D-L. Massart, and O. E. de Noord, “Comparison of multivariate calibration techniques applied to experimental NIR data sets,” Appl. Spectrosc. 54, 608–629 (2000). [CrossRef]
4. I. Georgakoudi, B. C. Jacobson, J. van Dam, V. Backman, M. B. Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A. Thomas, L. T. Perelman, and M. S. Feld, “Fluorescence, reflectance, and light-scattering spectroscopy for evaluating dysplasia in patients with Barrett’s esophagus,” Gastroenterology. 120, 1620–1629 (2001). [CrossRef] [PubMed]
5. T. Burger, J. Kuhn, R. Caps, and J. Fricke, “Quantitative determination of the scattering and absorption coefficients from diffuse reflectance and transmittance measurements,” J. Appl. Spectrosc. 51, 309–317 (1997). [CrossRef]
6. O. Berntsson, T. Burger, S. Folestad, L. G. Danielsson, J. Kuhn, and J. Fricke, “Effective sample size in diffuse reflectance near-IR spectrometry,” Anal. Chem. 71, 617–623 (1999). [CrossRef] [PubMed]
7. M. S. Patterson, B. Chance, and B. C. Wilson, “Time resolved reflectance and transmittance for the non-invasive measurement of optical properties,” Appl. Opt. 28, 2331–2336 (1989). [CrossRef] [PubMed]
8. M. S. Patterson, J. D. Moulton, B. C. Wilson, and B. Chance, “Applications of time-resolved light scattering measurements to photodynamic therapy dosimetry,” in Photodynamic Therapy: Mechanisms II, Proc. SPIE1205, 62–75 (1990).
9. S. J. Madsen, M. S. Patterson, B. C. Wilson, Y. D. Park, J. D. Moulton, S. L. Jacques, and Y. Hefetz, “Time resolved diffuse reflectance and transmittance studies in tissue simulating phantoms: a comparison between theory and experiment,” in Time-Resolved Spectroscopy and Imaging of Tissue B.Chance, ed. Proc. SPIE1431, 42–51 (1991).
10. S. Andersson-Engels, R. Berg, and S. Svanberg, “Effects of optical constants on time-gated transillumination of tissue and tissue-like media,” J. Photochem. Photobiol. B. 16, 155–167 (1992). [CrossRef] [PubMed]
11. S. Andersson-Engels, R. Berg, A. Persson, and S. Svanberg, “Multispectral tissue characterization with time-resolved detection of diffusely scattered white light,” Opt. Lett. 18, 1697–1699 (1993). [CrossRef] [PubMed]
12. R. Cubeddu, C. D’Andrea, A. Pifferi, P. Taroni, A. Torricelli, G. Valentini, M. Ruiz-Altisent, C. Valero, C. Ortiz, C. Dover, and D. Johnson, “Time-resolved reflectance spectroscopy applied to the nondestructive monitoring of the internal optical properties in apples,” Appl. Spectrosc. 55, 1368–1374 (2001). [CrossRef]
13. J. R. Lakowicz and K. Berndt, “Frequency-domain measurements of photon migration in tissues,” Chem. Phys. Lett. 166, 246 (1990). [CrossRef]
14. J. Fishkin, E. Gratton, M. J. vandeVen, and W. W. Mantulin, “Diffusion of intensity modulated near-infrared light in turbid media,” in Time-Resolved Spectroscopy and Imaging of TissueB. Chance, ed. Proc. SPIE1431, 122–135 (1991).
15. M. Patterson, J. D. Moulton, B. C. Wilson, K. W. Berndt, and J. R. Lakowicz, “Frequency-domain reflectance for the detemination of the scanttering and absorption properties of tissue,” Appl. Opt. 30, 4474–4476 (1991). [CrossRef] [PubMed]
16. S. J. Madsen, E. R. Anderson, R. C. Haskell, and B. J. Tromberg, “Portable, high-bandwidth frequency-domain photon migration instrument for tissue spectroscopy,” Opt. Lett. 19, 1934–1936 (1994). [CrossRef] [PubMed]
17. E. Gratton and J. Maier, “Frequency-domain measurements of photon migration in highly scattering media,” Medical Optical Tomography.534–544 (1996).
18. M. A. Franceschini, V. Toronov, M. E. Filiaci, E. Gratton, and S. Fantini, “On-line optical imaging of the human brain with 160-ms temporal resolution,” Opt. Express. 6, 49–57 (2000), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-6-3-49. [CrossRef] [PubMed]
19. F. Bevilacqua, A. J. Berger, A. E. Cerussi, D. Jakubowski, and B. J. Tromberg, “Broadband absorption spectroscopy in turbid media by combined frequency-domain and steady-state methods,” Appl. Opt. 39, 6498–6507 (2000). [CrossRef]
20. T. J. Farrell, B. C. Wilson, and M. S. Patterson, “The use of a neural network to determine tissue optical properties from spatially resolved diffuse reflectance measurements,” Phys. Med. Biol. 37, 2281–2286 (1992). [CrossRef] [PubMed]
21. J. S. Dam, C. B. Pedersen, T. Dalgaard, P. E. Fabricius, P. Aruna, and S. Andersson-Engels, “Fiber optic probe for non-invasive real-time determination of tissue optical properties at multiple wavelengths,” Appl. Opt. 40, 1155–1164 (2001). [CrossRef]
22. R. L. P. van Veen, W. Verkruysse, and H. J. C. M. Sterenborg, “Diffuse-reflectance spectroscopy from 500 to 1060 nm by correction for inhomogeneously distributed absorbers,” Opt. Lett. 27, 246–248 (2002). [CrossRef]
23. R. Cubeddu, A. Pifferi, P. Taroni, A. Torricelli, and G. Valentini, “Noninvasive absorption and scattering spectroscopy of bulk diffusive media: An application to the optical characterization of human breast,” Appl. Phys. Lett. 74, 874–876 (1999). [CrossRef]
24. A. Pifferi, J. Swartling, E. Chikoidze, A. Torricelli, P. Taroni, S. Andersson-Engels, and R. Cubeddu, “Spectroscopic time-resolved diffuse reflectance and transmittance measurements of the female breast at different interfiber distances,” J. Biomedical Optics. (to be published).
25. R. R. Alfano and S. L. Shapiro, “Observation of self-phase modulation and small-scale filaments in crystals and glasses,” Phys. Rev. Lett. 24, 592–594 (1970). [CrossRef]
27. O. Jarlman, R. Berg, S. Andersson-Engels, S. Svanberg, and H. Pettersson, “Time-resolved white light transillumination for optical imaging,” Acta Radiol. 38, 185–189 (1997). [PubMed]
28. J. Johansson, S. Folestad, M. Josefson, A. Sparen, C. Abrahamsson, S. Andersson-Engels, and S. Svanberg, “Time-resolved NIR/Vis spectroscopy for analysis of solids: Pharmaceutical tablets,” Appl. Spectrosc. 56, 725–731 (2002). [CrossRef]
29. J. K. Ranka, R. S. Windeler, and A. J. Stentz, “Visible continuum generation in air-silica microstructure optical fibers with anomalous dispersion at 800 nm,” Opt. Lett. 25, 25–27 (2000). [CrossRef]
30. J. C. Knight, T. A. Birks, R. F. Cregan, P. S. J. Russell, and J. P. de Sandro, “Photonic crystals as optical fibres - physics and applications,” Optical Materials. 11, 143–151 (1999). [CrossRef]
31. C. Abrahamsson, S. Andersson-Engels, S. Folestad, J. Johansson, and S. Svanberg. “New measuring technique”, Patent Application PCT WO 2002075286 (2002)
32. G. Genty, M. Lehtonen, H. Ludvigsen, J. Broeng, and M. Kaivola, “Spectral broadening of femtosecond pulses into continuum radiation in microstructured fibers,” Opt. Express. 10, 1083–1098 (2002), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-10-20-1083. [CrossRef] [PubMed]
33. J. Swartling, J. S. Dam, and S. Andersson-Engels, “Comparison of spatially and temporally resolved diffuse-reflectance measurement systems for determination of biomedical optical properties,” Appl. Opt. 42, 4612–4620 (2003). [CrossRef] [PubMed]
34. R. C. Haskell, L. O. Svaasand, T.-T. Tsay, T.-C. Feng, M. S. McAdams, and B. J. Tromberg, “Boundary conditions for the diffusion equation in radiative transfer,” J. Opt. Soc. Am. A. 11, 2727–2741 (1994). [CrossRef]
36. C. Abrahamsson, J. Johansson, A. Sparén, and F. Lindgren, “Comparison of different variable selection methods conducted on NIR transmission measurements on intact tablets,” Chemom. Intell. Lab. Syst. 69, 3–12 (2003). [CrossRef]