GO TO THE FULL PROTOCOL:
PDF or HTML at Wiley Online Library
Approaches to Spectral Imaging Hardware
Publication Name:
Current Protocols in Cytometry
Unit Number:
Unit 12.20
DOI:
10.1002/0471142956.cy1220s53
Online Posting Date:
July, 2010
Abstract
Instruments used for spectral, multispectral, and hyperspectral imaging in the biosciences have evolved significantly over the last 15 years. However, very few are calibrated and have had their performance validated. Now that spectral imaging systems are appearing in clinics and pathology laboratories, there is a growing need for calibration and validation according to universal standards. In addition, some systems produce spectral artifacts that, at the very least, challenge data integrity if left unrecognized. This unit includes a comparison of the band-pass and light-transmission characteristics of electronic tunable filters, interferometers, and wavelength-dispersive spectral imaging instruments, as well as a description of how they work. Methods are described to test wavelength accuracy and perform radiometric calibration. A real-life example of spectral artifacts is dissected in detail in order to show how to detect, diagnose, verify, and work around their presence when they cannot be eliminated. Curr. Protoc. Cytom. 53:12.20.1-12.20.40. © 2009 by John Wiley & Sons, Inc.
Keywords: acousto optic tunable filter; liquid crystal tunable filter; imaging spectrometer; prism spectrometer; diffraction grating ghosts; diffraction grating spectrometer; holographic diffraction grating; ruled diffraction grating
Table of Contents
- Introduction
- Spectral Imaging Basics
- Wavelength-Dispersive Spectral Imaging (WDSI)
- “Need to Know” Diffraction Grating Characteristics
- Electronic Tunable Filters (ETF)
- System-Based Light Throughput Considerations
- Wavelength Detectors
- Calibration, Validation, and Radiometry
- Comparison of Spectral Imaging Systems
- Acknowledgements
- Contact Information
- Literature Cited
- Figures
- Tables
Figures
-
Figure 12.20.1 (A) A nanowire internalized within a living cell as seen in polarized light, in transmission. (B) A spectral image of the nanowire in panel A; the spectrum confirms the presence of gold and nickel. -
Figure 12.20.2 The spectral “lines” from a MIDL wavelength calibration lamp are shown in the top panel. The bottom panel shows the spectrum; each line is an image of the entrance slit of the spectrometer. -
Figure 12.20.3 (A) A non-imaging spectrometer suffers from astigmatism resulting in a “point” on the entrance slit being imaged as “line” in the exit plane. (B) An imaging spectrometer is able to image a “point” on the entrance slit as a “point” on the detector at every wavelength. -
Figure 12.20.4 (A) An imaging Czerny-Turner spectrograph uses a toroidal mirror to correct astigmatism with either a classically ruled, or holographic, plane grating. (B) Offner imaging spectrograph uses either classically ruled, or holographic, convex grating. (C) Single-element, imaging aberration reduced, concave holographic grating spectrograph. -
Figure 12.20.5 Prism-based imaging spectrograph employs a prism with curved sides to correct astigmatism and other aberrations over a wide FOV. -
Figure 12.20.6 Refractive index and wavelength dispersion curves for a flint prism. Band-pass decreases nonlinearly with wavelength. -
Figure 12.20.7 Spectral resolution is measured at the full width at half maximum of an “infinitely” narrow emission line. In this case, the 405-nm Hg line is resolved with an optimum of three camera pixels. -
Figure 12.20.8 The diffraction efficiency profiles of a ruled and a holographic grating present discontinuities in their efficiency profiles. The internal transmission curve of a flint glass prism offers a flat response from 365 nm to 950 nm. -
Figure 12.20.9 (A) A 600 g/mm diffraction grating distributes light at 532 nm into 3 positive and 3 negative orders. (B) A 150 g/mm diffraction grating distributes light at 532 nm into 12 positive and 12 negative orders. -
Figure 12.20.10 (A) A wavelength-calibration lamp spectrum showing second-order superimposed on the first-order spectrum when using a diffraction grating–based spectrometer. (B) A wavelength-calibration lamp spectrum using a prism-based spectrometer shows no second-order contamination. -
Figure 12.20.11 (A) The positive and negative first orders of a 1200 g/mm ruled diffraction grating. Ghosts and focused stray light are visible between zero and plus and minus first order. (B) A 1200 g/mm holographic grating shows no ghosts or focused stray light. (C) A more extreme example of a 1200 g/mm grating presenting ghosts and focused stray light. -
Figure 12.20.12 (A) Cells excited at 405 nm in a spectral confocal system present an artifact at 494 nm. (B) Spectral confocal image of 561- and 488-nm lasers reflected off a front surface mirror; note the two artifacts at 688 and 707 nm. (C) Spectral characteristics of the 561 and 488 laser lines showing spectral artifacts. -
Figure 12.20.13 (A) Wavelength dispersion curves for a diffraction grating and a prism. In both cases, dispersion is nonlinear but far more pronounced for a prism than a diffraction grating. (B) Spectrum of a low-pressure Hg wavelength calibration lamp acquired on a PARISS prism-based imaging spectrometer. Lines from the Hg 365-nm line to the Ar 912-nm line are clearly resolved. -
Figure 12.20.14 (A) Schematic of an AOTF. Broadband unpolarized light enters from the left. An acoustic wave is launched with an RF signal to drive an apodized transducer. This wave sets up periodic modulations of the index of refraction in the crystal to diffract incoming light into orthogonally polarized +1 and –1 order beams. in is the half-angle of the input aperture of the crystal.0 is the cut of the input face of the crystal relative to the crystalline Z axis.exit is the angle of the exit face of the crystal; when properly chosen, this ensures that the +1 diffracted beam direction does not vary with wavelength. (B) Relative wavelength dispersion curves for an AOTF and LCTF. (C) Relative transmission curves for a LCTF and AOTF. (D) Transmission profile of an apodized AOTF. A non-apodized transducer produces a sinc2 transmission profile presenting significant side lobes. An apodized transducer eliminates those side lobes. -
Figure 12.20.15 (A) An LCTF is positioned in front of a camera; the FOV is imaged through the cell assembly. (B) An LCTF is composed of up to 10 individual cells. (C) Each cell consists of a nematic liquid crystal sandwiched between a polarizer and a polarization retarder waveplate. The wavelength transmitted through the cell is changed by applying a voltage to the nematic liquid crystal. -
Figure 12.20.16 (A) LCTF transmission is the product of the transmission of each cell in the assembly. (B) LCTF percent transmission bands in randomly polarized light; spectral overlap increases with wavelength. -
Figure 12.20.17 Layout of a Sagnac interferometer. An image of a fixed FOV is acquired at each wavelength sequentially when the Optical Path Difference is changed by rotating mirrors M1 and M2. -
Figure 12.20.18 (A) QE curves of a PMT and a scientific CCD camera. Quantum curves vary widely from PMT to PMT and camera to camera. (B) Relative intensity curves for halogen and xenon lamps. -
Figure 12.20.19 (A) The spectrum of an MIDL wavelength calibration lamp presenting both Hg and Ar lines. The Ar lines fade rapidly after the lamp is turned on, and the Hg lines increase in intensity. (B) Chart of wavelengths present in a MIDL lamp. -
Figure 12.20.20 (A) The power spectrum of a NIST-certified halogen lamp; this spectrum is used to develop a software correction curve to render an instrument “Radiometric.” (B) The raw spectrum of a NIST-certified halogen lamp is a convolution of the QE of the camera and the transmission characteristics of all other optics in the system. After the application of a correction factor, the raw curve is identical to the certified spectrum. (C) The power spectra of a xenon and halogen lamp after radiometric correction. -
Figure 12.20.21 (A) Efficiency curves of AOTF, LCTF, classically ruled and holographic diffraction gratings and a flint-glass prism compared. (B) Comparison of relative band-pass of AOTF, LCTF, classically ruled, and holographic diffraction gratings, and a flint glass prism.
Videos
Literature Cited
| Literature Cited | |
| Cabib, D., Buckwald, R.A., and Nissim, B.-Y. 1998. Method for chromosome classification by decorrelation statistical analysis and hardware. U.S. Patent No. 5,719,0124. | |
| Chrisp, M.P. 1987. "Aberration-corrected holographic gratings and their mountings". In Applied Optics and Optical Engineering (R.R. Shannon and W.C. Wyant, eds.) pp. 391-451. Academic Press, London | |
| Dicker, D.T., Lerner, J.M., Van Belle, P., Barth, S.F., Guerry, D., Herlyn, M., Elder, D.E. and El-Deiry, W.S. 2006. Differentiation of normal skin and melanoma using high resolution hyperspectral imaging. Cancer Biol. Ther. 8:1033-1038. | |
| Dickinson, M., Bearman, G., Tille, S., Lansford, R., and Fraser, S. 2001. Multi-spectral imaging and linear unmixing add a whole new dimension to laser scanning fluorescence microscopy. BioTechniques 31:1272-1278. | |
| Flamand, J., Bonnemason, F., Lerner, J.M., and Thevenon, A. 1989. Blazing of holographic gratings using ion-etching. Proc. SPIE 1055:288. | |
| Garini, Y., Katzir, N., Cabib, D., Buckwald, R., Soenksen, D., and Malik, Z. 1996. "Fluorescence imaging spectroscopy and microscopy". In Fluorescence Imaging Spectroscopy and Microscopy (X. Wang and B. Herman, eds.). pp. 87-124. John Wiley & Sons, New York. | |
| Gat, N. 2000. Imaging spectroscopy using tunable filters: A review. Proc. SPIE 4056:50-64. | |
| Hutley, M.C. 1982. Diffraction Gratings. Academic Press, London. | |
| James, J. and Sternberg, R. 1969. The Design of Optical Spectrometers. Chapman & Hall, London. | |
| Lerner, J.M. 2006. Imaging spectrometer fundamentals for researchers in the biosciences: A tutorial. Cytometry A 69:712-734. | |
| Lerner, J. and Thevenon, A. 1998. The Optics of Spectroscopy: A Tutorial. Horiba, Metuchen, New Jersey. | |
| Lerner, J.M. and Zucker, R. 2004. Calibration and validation of confocal spectral imaging systems. Cytometry A 62:8-34. | |
| Loewen, E. and Popov, E. 1997. Diffraction Gratings and Applications. Marcel Dekker, New York. | |
| Miller, P.J. 1991. Use of tunable liquid crystal filters to link radiometric and photometric standards. Metrologia 28:145-149. | |
| Prieto-Blanco, X., Montero-Orille, C., Couce, B., and de la Fuente, R. 2006. Analytical design of an Offner imaging spectrometer. Opt. Express 14:9156-9168. | |
| Reader, J. 1969. Optimizing Czerny-Turner spectrographs: A comparison between analytical theory and ray tracing. J. Opt. Soc. Am. 59:1189-1196. | |
| Shabtay, G., Eidinger, E., Zalevsky, Z., Mendlovic, D., and Marom, E. 2002. Tunable birefringent filters: Optimal iterative design. Opt. Express 10:1534-1541. | |
| Stroke, G.W. 1967. "Diffraction gratings (in English)". In Handbuch der Physik XXIX (S. Flügge, ed.), vol. 29. Springer-Verlag, Berlin. | |
| Tumeh, P.C., Lerner, J.M., Dicker, D.T., and El-Deiry, W.S. 2007. Differentiation of vascular and non-vascular skin spectral signatures using in vivo hyperspectral radiometric imaging: Implications for monitoring angiogenesis. Cancer Biol. Ther. 6:447-453. | |
| van der Loos, C.M. 2008. Multiple immunoenzyme staining: Methods and visualizations for the observation with spectral imaging. J. Immunochem. Cytochem. 56:313-328. | |
| Vo-Dinh, T. 2004. A hyperspectral imaging system for in vivo optical diagnostics. IEEE Engin. Med. Biol. Mag. 23:40-49. | |
| Wachman, E.S. and Pannell, C.N. 2008. High-performance hyperspectral imager using a novel acousto-optic tuneable filter. Proc. SPIE 6966:7-12. | |
| Warren, D., Hackwell, J., and Gutierrez, D. 1997. Compact prism spectrographs based on aplanatic principles. Opt. Eng. 36:1174-1182. | |
| Ye, C. 2003. Wavelength-tunable spectral filters based on the optical rotatory dispersion effect. Appl. Optics 42:4505-4513. | |
