ENDOSCOPY: Coherent Raman scanning-fiber endoscope images at 7 fps
Both stimulated Raman scattering (SRS) and coherent anti-Stokes Raman scattering (CARS) microscopy use light from a pump and probe laser beam scanned over tissue or a test sample to accomplish chemically selective, label-free, high-spatial-resolution imaging at up to video rates.
Both stimulated Raman scattering (SRS) and coherent anti-Stokes Raman scattering (CARS) microscopy use light from a pump and probe laser beam scanned over tissue or a test sample to accomplish chemically selective, label-free, high-spatial-resolution imaging at up to video rates. These coherent Raman laser-scanning microscopes were not able to be miniaturized easily for endoscopy. But researchers at Harvard University (Cambridge, MA) and the University of Washington (Seattle, WA) have succeeded in developing a coherent Raman scanning-fiber endoscope, enabling SRS and CARS microscopy for in situ biological tissue analysis.1
In the standard coherent Raman scattering test setup, a 1064 nm neodymium:yttrium vanadate (Nd:YV04) laser provides 80 MHz repetition rate, 7 ps Stokes (probe-beam) pulses. A portion of this output is frequency doubled and sent to an optical parametric oscillator (OPO) to produce the pump beam at 800 nm. The Stokes beam is modulated and combined with the pump beam via a dichroic mirror. The beams are then linearly polarized and properly oriented for delivery along the proper axis of a 1-m-long section of polarization-maintaining (PM) optical fiber.
The fiber tip is driven at its mechanical resonance frequency in an expanding spiral by a piezoelectric actuator, enabling a scan rate of 7 frames per second (fps). The scan pattern is created by inserting the tip of the PM fiber through a 0.5-mm-diameter, piezoelectrically active ceramic tube radially patterned with four electrodes (see figure). With one pair of opposing electrodes driven with a sine wave and the other driven with a cosine wave (both with linearly expanding envelope functions), a spiral scan pattern is achieved. The driver electronics track the fiber tip for image reconstruction as the data is processed.
The researchers avoided the typical axial chromatic aberrations of a single-element gradient-index (GRIN) lens by using a chromatically corrected hybrid lens consisting of two GRIN lenses sandwiching a diffractive lens to focus the excitation light. The 1.4-mm-diameter lens uses the diffractive optic to achieve near-zero axial chromatic aberration at the chosen wavelengths. Because the sources emit picosecond pulses, ordinary silica fibers can be used for light delivery over 1 m without significant self-phase-modulation broadening of the optical spectrum beyond the intrinsic Raman linewidth (approximately 1 nm at 800 nm).
|A piezoelectrically (PZ) actuated scanning-fiber endoscope delivers pump/Stokes light for CARS and SRS microscopy in a spiral pattern to biological tissue through a special gradient-index (GRIN) objective lens (OL). An optical filter (OF) transmits the anti-Stokes wavelength to a photodiode (PD) for processing. The physical endoscope uses a 1 m section of fiber and the fiber scanner and objective lens are only 1.4 mm in diameter. Though the current system uses a 10 mm photodiode for light collection, simulations indicate that collection using a 5 mm or smaller detector will be possible. (Courtesy of Harvard University)|
Experimental imaging of 2-µm-diameter polystyrene beads in an 80-µm-diameter field of view yielded axial full-width at half-maximum (FWHM) resolution of 6.5 µm and lateral FWHM resolution of 1.3 µm. Imaging of mouse tissue and hair confirmed the SRS scanning-fiber endoscope system was able to distinguish the chemical contrast between lipids and proteins at depths up to 30–50 µm into tissue. The contrast is comparable to previous CARS and SRS measurements. “There has been a lot of interest in developing miniaturized coherent Raman systems for medical applications in vivo, because the unique combination of label-free chemical contrast and microscopic resolution shows great promise for distinguishing healthy tissue from cancerous tumors, for example,” says Brian Saar, lead author of the paper and a graduate student at Harvard University when this work was done. —Gail Overton
1. B.G. Saar et al., Opt. Lett., 36, 13, 2396–2398 (2011).