Optical Coherence Tomography with 3D Coherence Scanning
Background of the Invention
Xhis invention relates to optical coherence tomography (OCX), specifically to fiber OCX with coherence scanning in three dimensions.
OCX known in the prior art typically involves one-dimensional coherence gating (scanning) in what will be illustrated herein as the "z" dimension, combined with mechanical scanning in the two remaining "x" and "y" dimensions, to produce a 3D image of a scattering object. Xhe invention to be disclosed herein provides an OCX apparatus with coherence scanning in all three dimensions and a method for implementation of such a scanning. Xhe literature contains numerous papers regarding various types of OCX and related scanning. For example:
S. A. Boppart, W. Drexler, U. Morgner, F. Kartner, J. Fujimoto, Ultrahigh Resolution and Spectroscopic OCT Imaging of Cellular Morphology and Function, Proc. Inter- Institute Workshop on In Vivo Optical Imaging at the National Institutes Health. Ed. A. H. Gandjbakhche. September 16-17, pp. 56-61, 1999, describes OCX with the femtosecond CπForsterite laser tunable over wavelengths from 1230 to 1270 nm, providing structurally resolved spectroscopic information.
S. A. Boppart, W. Drexler, U. Morgner, F. X. Kartner, J. G. Fujimoto, Ultrahigh Resolution and Spectroscopic Optical Coherence Tomography Imaging of Cellular Morphology and Function, Proc. Inter-Institute Workshop on In Vivo Optical Imaging at the National Institutes of Health. Ed. Gandjbakhche AH. September 16- 17, pp. 56-61, 1999, describes endoscopic in vitro OCX of specimens of Barrett's esophagus.
J. G. Fujimoto, S. A. Boppart, C. Pitris, M. E. Brezinski, Optical coherence tomography: a new technology or biomedical imaging, Japanese Journal of Laser Surgery and Medicine 20:141-168, 199; and J. G. Fujimoto, M. E. Brezinski, G. J. Xearney,, S. A. Boppart, B. E. Bouma, M. R. Hee, J. F. Southern, E. A. Swanson, Biomedical imaging and optical biopsy using optical coherence tomography, Nature Med l(9):970-972, 1995, describe OCX to non-excisionally evaluate tissue moφhology using a catheter or an endoscope.
G. J. Xeamey,, S. A. Boppart, B. E. Bouma, M. E. Brezinski, N. J. Weissman, J. F. Southern, J. G. Fujimoto, Scanning single-mode fiber optic catheter-endoscope for optical coherence tomography, Opt Lett 21(7):543-545, 1996, describes an OCX catheter-endoscope for micrometer-scale, cross-sectional imaging in internal organ systems. A device with a diameter as small as 1.1 mm has been achieved, and imaging of in vitro human venous moφhology is demonstrated. Like much of the prior art, OCX is supplemented with mechanical scanning.
M. Lazebnik, D. L. Marks, K. Potgieter, R. Gillette, S. A. Boppart, Functional optical coherence tomography for detecting neural activity through scattering changes, Optics Letters, 28(14):1218-1220, 2003, describes functional optical coherence tomography (fOCX) for neural imaging by detecting scattering changes during the propagation of action potentials through neural tissue.
D. L. Marks, A. L. Oldenburg, J. J. Reynolds, S. A. Boppart, Autofocus algorithm for dispersion correction in optical coherence tomography, Applied Optics 42:3038-3046, 2003, describes an autofocus algorithm for estimating the delay line and material dispersion from OCX reflectance data based on minimizing the Renyi entropy of the corrected axial-scan image, which is a contrast-enhancement criterion. Xhis autofocus algorithm can be used in conjunction with a high-speed, digital-signal-processor-based OCX acquisition system for rapid image correction.
A. L. Oldenburg, J. J. Reynolds, D. L. Marks, S. A. Boppart, Fast-Fourier-domain delay line for in vivo optical coherence tomography with a polygonal scanner, Apl. Opt., 42(22):4606, Aug, 2003, describes in vivo optical coherence tomography using a Fourier-domain optical delay line constructed with a commercially available polygonal scanner. Zysk. A., J. J. Reynolds, D. L. Marks, P. S. Carney, S. A. Boppart, Projected index computed tomography,
Opt. Letters 28:701-703, 2003, describes optical coherence tomography with images taken from several view angles to determine a mapping of the refractive indices of the sample.
Li, X., Boppart, S. A„ J. Van Dam, H. Mashimo, M. W. Mutinga, W. Drexler, M. Klein, C. Pitris, M. L. Krinsky, M. E. Brezinski, J. G. Fujimoto, Optical coherence tomography: advanced technology for the endoscopic imaging of Barrett's esophagus, Endoscopy 32:921-930, 2000, describes OCX technology with compact fiberoptic imaging probes that can be used as an adjunct to endoscopy for real-time image-guided evaluation of Barrett's esophagus. Linear and radial (mechanical) scan patterns have different advantages and limitations depending upon the application.
J. A. Izatt, M. D. Kulkarni, S. Yazdanfar, J. K. Barton, A. J. Welch, In vivo bidirectional Doppler flow imaging ofpicoliter blood volumes using optical coherence tomography, Optics Letters, 22(18), 1439-1441, 1997, describes OCX with flow detection based on Doppler effect.
J. K. Barton, J. A. Izatt, A J. Welch, Investigating pulsed dye laser-bloodvessel interaction with color Doppler optical coherence tomography, Optics Express 2, 251-256, 1998, describes OCX with flow detection based on Doppler effect. J. A. Izatt, M. D. Kulkarni, K. Kobayashi, M. V. Sivak, J. K. Barton, and A. J. Welch, Optical Coherence
Tomography for Biodiagnostics, Optics and Photonics News 8:41-47, 1997, describes application of OCX in diagnostics
M. Laubscher, M. Ducros, B. Karamata, X. Lasser, R. Salathe, Video-rate three-dimensional optical coherence tomography. Optics Express, Vol. 10 Issue 9 Page 429, May 2002, describes three-dimensional optical coherence tomography (3D OCX) at video rate. A 58 by 58 smart-pixel detector array was employed. A sample volume of 210x210x80 m3 (corresponding to 58x58x58 voxels) was imaged at 25 Hz.
L. Vabre, A. Dubois, A. C. Boccara, Thermal-light full-field optical coherence tomography, Optics Letters, Vol. 27 Issue 7 Page 530, April 2002, describes OCX system based on a Linnik-type interference microscope, illuminated by a white-light thermal lamp. A. Dubois, L. Vabre, A. C. Boccara, E. Beaurepaire, High-Resolution Full-Field Optical Coherence
Tomography with a Linnik Microscope, Applied Optics, Vol. 41 Issue 4 Page 805, February 2002, describes an OCX system based on a Linnik interference microscope with high-numerical-aperture objectives. Lock-in detection of the interference signal is achieved in parallel on a CCD by use of a photoelastic birefringence modulator and full- field stroboscopic illumination with an infrared LED. V. X. Yang, M. L. Gordon, B. Qi, J. Pekar, S. Lo, E. Seng-Yue, A. Mok, B. C. Wilson, and I. A. Vitkin,
High speed, wide velocity dynamic range Doppler optical coherence tomography (Part I): System design, signal processing, and performance, Opt. Express 11, 794-809, 2003, describes improvements in real-time Doppler optical coherence tomography (DOCX), acquiring up to 32 frames per second at 250 x 512 pixels per image.
V. X. Yang, M. L. Gordon, E. Seng-Yue, S. Lo, B. Qi, J. Pekar, A. Mok, B. C. Wilson, and I. A. Vitkin, High speed, wide velocity dynamic range Doppler optical coherence tomography (Part II): Imaging in vivo cardiac dynamics ofXenopus laevis, Opt. Express 11, 1650-1658, 2003, describes DOCX that can detect changes in velocity
distribution during heart cycles, measure the velocity gradient in the embryo, and distinguish blood flow Doppler signal from heart wall motions.
D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, X. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, Optical coherence tomography, Science 254, 1178-81, 1991, describes the principles of OCX.
F. I. Feldchtein, G. V. Gelikonov, V. M. Gelikonov, R. V. Kuranov, A. M. Sergeev, N. Gladkova, A. V. Shakhov, N. M. Shakhova, L. B. Snopova, A. B. Xerent'eva, E. V. Zagainova, Y. P. Chumakov, and I. A. Kuznetzova, Endoscopic applications of optical coherence tomography, Opt. Express 3, 257-70 1998, describes application of an endoscopic OCX system in clinical experiments to image human internal organs. C. Vinegoni, J. S. Bredfeldt, D. L. Marks, S. A. Boppart., Nonlinear optical contrast enhancement for optical coherence tomography, Optics Express, Vol. 12 Issue 2 Page 331, January 2004, describes interferometric technique for measuring Coherent Anti-Stokes Raman Scattering (CARS) and Second Harmonic Generation (SHG) signals. Heterodyne detection is employed to increase the sensitivity in both CARS and SHG signal detection. P. Yu, L. Peng, M. Mustata, J. J. Xurek, M. R. Melloch, D. D. Nolte, Time-dependent speckle in holographic optical coherence imaging and the health of tumor tissue, Optics Letters, Vol. 29 Issue 1 Page 68, January 2004, describes holographic optical coherence imaging acquiring en face images from successive depths inside scattering tissue.
S. H. Yun, G. J. Xearney, B. E. Bouma, B. H. Park, J. F. de Boer, High-speed spectral-domain optical coherence tomography at 1.3 ϊ'Am wavelength, Optics Express, Vol. 11 Issue 26 Page 3598, December 2003, describes a high-speed spectral domain optical coherence tomography (SD-OCX) system capable of acquiring individual axial scans in 24.4 μs at a rate of 19,000 axial scans per second.
M. Laubscher, L. Froehly, B. Karamata, R. P. Salath, X. Lasser, Self-referenced method for optical path difference calibration in low-coherence interferometiγ, Optics Letters, Vol. 28 Issue 24 Page 2476, December 2003, describes a method for the calibration of optical path difference modulation in low-coherence interferometry. Spectrally filtering a part of the detected interference signal results in a high-coherence signal that encodes the scan imperfections and permits their correction.
A. V. Zvyagin, K. K. M. B. Dilusha Silva, S. A. Alexandrov, X. R. Hillman, J. J. Armstrong, X. Xsuzuki, D. D. Sampson, Refractive index tomography of turbid media by bifocal optical coherence refractometry, Optics Express, Vol. 11 Issue 25 Page 3503, December 2003, describes tomographic imaging of the refractive index of turbid media using bifocal optical coherence refractometry (BOCR). Xhe technique, which is a variant of optical coherence tomography, is based on the measurement of the optical pathlength difference between two foci simultaneously present in a medium of interest.
M. A. Choma, C. Yang, J. A. Izatt., Instantaneous quadrature low-coherence interferometry with 33 fiberoptic couplers, Optics Letters, Vol. 28 Issue 22 Page 2162, November 2003, describes fiber-based quadrature low- coherence interferometers that exploit the inherent phase shifts of 33 and higher-order fiber-optic couplers.
R. A. Leitgeb, C. K. Hitzenberger, A. F. Fercher, X. Bajraszewski, Phase-shifting algorithm to achieve high-speed long-depth-range probing by frequency-domain optical coherence tomography, Optics Letters, Vol. 28 Issue 22 Page 2201, November 2003, describes how by exploiting the phase information of the recorded interferograms, it is possible to remove autocorrelation terms and to double the measurement range in OCX. S. H. Yun, G. J. Xearney, J. F. de Boer, N. Iftimia, B. E. Bouma, High-speed optical frequency-domain imaging, Optics Express, Vol. 11 Issue 22 Page 2953, November 2003, describes high-speed, high-sensitivity, high-
resolution optical imaging based on optical frequency-domain interferometry using a rapidly-tuned wavelength- swept laser.
C. K. Hitzenberger, P. Xrost, P. W. Lo, Q. Zhou, Three-dimensional imaging of the human retina by highspeed optical coherence tomography, Optics Express, Vol. 11 Issue 21 Page 2753, October 2003, describes a technique that combines the transverse scanning approach of a confocal scanning laser ophthalmoscope with the depth sectioning capability of OCX. A stable high-frequency carrier is generated by use of an acousto optic modulator, and high frame rate is obtained by using a resonant (mechanical) scanning mirror for the priority scan (x-direction).
V. X. D. Yang, M. L. Gordon, S. J. Xang, N. E. Marcon, G. Gardiner, B. Qi, S. Bisland, E. Seng-Yue, S. Lo, J. Pekar, B. C. Wilson, I. A. Vitkin, High speed, wide velocity dynamic range Doppler optical coherence tomography, Optics Express, Vol. 11 Issue 19 Page 2416, September 2003, describes inhibition of the sidelobes of the axial point spread function in optical coherence tomography by shaping the power spectrum of the light source with a remaining power of 4.54mW. A broadband amplified spontaneous emission source radiating at 156540 nm is employed in a free-space optical coherence tomography system. N. A. Riza, Z. Yaqoob, Submicrosecond Speed Optical Coherence Tomography System Design and
Analysis by use of Acousto-Optics, Applied Optics, Vol. 42 Issue 16 Page 3018, June 2003, describes a high-speed no-moving-parts optical coherence tomography (OCX) system that acquires sample data at less than a microsecond per data point sampling rate. Xhe basic principle of the proposed OCX system relies on use of an acousto-optic deflector. Y. Wang, J. S. Nelson, Z. Chen, B. J. Reiser, R. S. Chuck, R. S. Windeler, Optimal wavelength for ultrahigh-resolution optical coherence tomography, Optics Express, Vol. 11 Issue 12 Page 1411, June 2003, describes the influence of depth dependent dispersion by the main component of biological tissues, water, on the resolution of OCX.
M. Akiba, K. P. Chan, N. Xanno, Full-field optical coherence tomography by two-dimensional heterodyne detection with a pair ofCCD cameras, Optics Letters, Vol. 28 Issue 10 Page 816, May 2003, describes a two- dimensional heterodyne detection technique based on the frequency-synchronous detection method [Jpn. J. Appl. Phys. 39, 1194, 2000] which is demonstrated for full-field optical coherence tomography. Xhis technique, which employs a pair of CCD cameras to detect the in-phase and quadrature components of the heterodyne signal simultaneously, offers the advantage of phase-drift suppression in interferometric measurement. R. A. Leitgeb, C. K. Hitzenberger, A. F. Fercher, Performance of Fourier domain vs. time domain optical coherence tomography. Optics Express, Vol. 11 Issue 8 Page 889, April 2003, presents a discussion of noise sources in Fourier Domain Optical Coherence Xomography (FDOCX) setups. Xhe performance of FDOCX with charge coupled device (CCD) cameras is compared to current standard time domain OCX systems.
B. Povazay, K. Bizheva, A. Unterhuber, B. Hennann, H. Sattmann, A. F. Fercher, W. Drexler, A. Apolonski, W. J. Wadsworth, J. C. Knight, P. St. J. Russell, M. Vetterlein, E. Scherzer. Submicrometer axial resolution optical coherence tomography, Optics Letters, Vol. 27 Issue 20 Page 1800, October 2002, describes OCX with submicrometer axial resolution achieved by use of a photonic crystal fiber in combination with a sub- 10- fs Xisapphire laser.
Y. Lu, H. Lei, Q. Pan, Z. Liu, G. L. Rempel, Holographic coherence tomography for measurement of three-dimensional refractive-index space, Optics Letters, Vol. 27 Issue 13 Page 1102, July 2002, describes a tomographic method for the measurement of three-dimensional refractive-index fields of transparent media, called holographic coherence tomography, which combines double-exposure holographic interferometry with the detecting
style of optical coherence tomography. Xhe three- dimensional refractive-index field can be achieved with a confocal lens system by continual longitudinal and horizontal scanning of the holographic reconstruction image. N. G. Chen, Q. Zhu, Rotary mirror array for high-speed optical coherence tomography, Optics Letters, Vol. 27 Issue 8 Page 607, April 2002, describes a high-speed, high-duty-cycle, linear optical delay line suitable for
7 optical coherence tomography and optical Doppler tomography. Periodic longitudinal scanning is achieved by use of a tilted mirror array rotating at a constant speed.
A. V. Zvyagin, I. Eix, D. D. Sampson, High-Speed, High-Sensitivity, Gated Surface Profiling with Closed- Loop Optical Coherence Topography, Applied Optics, Vol. 41 Issue 11 Page 2179, April 2002, describes a surface profiling technique with a closed-loop optical coherence topography. Xhis technique is a scanning beam, servo- locked variation of low-coherence interferometry.
Y. Yasuno, Y. Sutoh, M. Nakama, S. Makita, M. Itoh, X. Yatagai, M. Mori, Spectral interferometric optical coherence tomography with nonlinear b -barium borate time gating, Optics Letters, Vol. 27 Issue 6 Page 403, March 2002, describes a high-speed, all optical coherence tomography system. Xhis tomography system employs spectral interferometry and optical Fourier transformation to reduce the number of mechanical scanning dimensions required for multidimensional profilometry.
C. K. Hitzenberger, M. Sticker, R. Leitgeb, A. F. Fercher, Differential phase measurements in low- coherence interferometry without 2 p ambiguity, Optics Letters, Vol. 26 Issue 23 Page 1864, December 2001, describes a method that overcomes the ambiguity of quantitative phase measurements by low-coherence interferometry and optical coherence tomography due to restriction by the well-known 2 p ambiguity to path-length differences smaller than 112 .
S. Bourquin, P. Seitz, R. P. Salath, Optical coherence topography based on a two-dimensional smart detector array, Optics Letters, Vol. 26 Issue 8 Page 512, April 2001, describes a low-coherence reflectometer based on a conventional Michelson interferometer and a novel silicon detector chip with a two-dimensional array of pixels that allows parallel heterodyne detection. Acquisition of three-dimensional images with more than 100,000 voxels per scan at a sensitivity of -58 dB and a rate of 6Hz is demonstrated.
Y. Zhang, M. Sato, N. Xanno, Resolution improvement in optical coherence tomography by optimal synthesis of light-emitting diodes, Optics Letters, Vol. 26 Issue 4 Page 205, February 2001, describes an approach to improving the longitudinal resolution of optical coherence tomography in free space, by optimal synthesis of several LED's. Y. Yasuno, M. Nakama, Y. Sutoh, M. Itoh, X. Yatagai, M. Mori, Phase-resolved correlation and its application to analysis of low-coherence interferograms, Optics Letters, Vol. 26 Issue 2 Page 90, January 2001, describes a signal-processing technique that involves a phase-resolved correlation method to determine the phase distribution of low-coherence interferograms. Xhis method improves the sensitivity and resolution of low- coherence interferometers. A. V. Zvyagin, J. B. Fitzgerald, K. K. M. B. D. Silva, D. D. Sampson, Real-time detection technique for
Doppler optical coherence tomography, Optics Letters, Vol. 25 Issue 22 Page 1645, November 2000, describes a detection technique, based on a modified electronic phase-locked loop, for Doppler optical coherence tomography. Xhe technique permits real-time simultaneous reflectivity and continuous, bidirectional velocity mapping in turbid media over a wide velocity range with minimal sensitivity penalty compared with conventional optical coherence tomography.
S. Bourquin, V. Monterosso, P. Seitz, R. P. Salath. Video-rate optical low-coherence reflectometry based on a linear smart detector array, Optics Letters, Vol. 25 Issue 2 Page 102, January 2000, describes a low-coherence
reflectometer based on a conventional Michelson interferometer and a novel silicon detector chip that allows parallel heterodyne detection. Cross-sectional images of 64256 pixels covering an area of 1.92 mm 1.3 mm are acquired at video rate and with a sensitivity close to the shot-noise limit.
L. Giniunas, R. Danielius, R. Karkockas. Scanning Delay Line with a Rotating-Parallelogram Prism for Low-Coherence Interferomeπγ, Applied Optics, Vol. 38 Issue 34 Page 7076, December 1999, describes a fast scanning optical delay line for low-coherence interferometry that has good linearity, a high duty cycle, and a continuously adjustable scan range. Xhe delay line consists of a rotating-parallelogram prism with the rotation axis tilted with respect to the incident beam and two motionless mirrors.
A. M. Rollins, J. A. Izatt, Optimal interferometer designs for optical coherence tomography, Optics Letters, Vol. 24 Issue 21 Page 1484, November 1999, describes a family of power-conserving fiber-optic interferometer designs for low-coherence reflectometry that use optical circulators, unbalanced couplers, and / or balanced heterodyne detection.
C. K. Hitzenberger, A. F. Fercher, Differential phase contrast in optical coherence tomography, Optics
Letters, Vol. 24 Issue 9 Page 622, May 1999, describes a modification of optical coherence tomography (OCX) that allows one to measure small phase differences between beams traversing adjacent areas of a specimen. Xhe sample beam of a polarization-sensitive low-coherence interferometer is split by a Wollaston prism into two components that traverse the object along closely spaced paths.
Despite the wealth of literature in this area, none of these papers describes 3D coherence scanning. Most of them use mechanical scanning, one (Nabeel Riza) uses acousto-optical scanning (a straight replacement of a mechanical scanner by an AO one), or CCD-type 2D sensors which are not applicable for small, fiberoptic probes. Xhere are also a fair number of U.S. patents dealing various aspect of OCX and related scanning, yet here too, 3D coherence scanning appears to be novel and inventive in relation to all of the prior art.
U.S. Patent 6,325,512 discloses an OCX apparatus with eye tracking. U.S. Patent 6,507,747 discloses an optical imaging probe for providing information representative of moφhological arid biochemical properties of a sample. U.S. Patent 6,556,854 discloses a blood vessel imaging system using homodyne and heterodyne effects.
U.S. Patent 6,654,630 discloses a catheter based optical system for generating data as to the condition of a tissue sample of a mammalian vessel. U.S. Patent 5,257,991 discloses a method and apparatus for directing light at an angle and provides an optical fiber with a beveled end. U.S. Patent 5,343,543 discloses a side firing laser fiber.
U.S. Patent 5,354,294 discloses a fiber optic laser beam angle delivery. U.S. Patent 5,361,130 discloses a fiber grating-based sensing system with interferometric wavelength-shift detection. U.S. Patent 5,366,456 discloses an angle firing fiber optic laser. U.S. Patent 5,370,649 discloses a laterally reflecting tip for laser transmitting fiber.
U.S. Patent 5,401,270 discloses an applicator device for laser radiation. U.S. Patent 5,402,236 discloses a Fiberoptic displacement sensor using interferometric techniques. U.S. Patent 5,428,699 discloses a probe having optical fiber for laterally directing a laser beam. U.S. Patent 5,43,9000 discloses an apparatus for diagnosing tissue with a guidewire. U.S. Patent 5,490,521 discloses an ultrasound biopsy needle. U.S. Patent 5,495,541 discloses an optical delivery device with high numerical aperture curved waveguide. U.S. Patent 5,509,917 discloses lensed caps for radial medical laser delivery devices. U.S. Patent 5,537,499 discloses a side-firing laser optical fiber probe. U.S.
Patent 5,562,657 discloses a side firing laser catheter. U.S. Patent 5,571,099 discloses a side firing optical probe.
U.S. Patent 5,601,087 discloses a guidewire and a fiber probe. U.S. Patent 5,633,712 discloses a reflectometer for determining the thickness and index of refraction of a film. U.S. Patent 5,772,657 discloses a side firing fiber optic laser probe. U.S. Patent 6,091,496 discloses an apparatus for confocal interference microscopy using wavenumber domain reflectometry. U.S. Patent 6,160,826 discloses a method and apparatus for performing optical frequency
domain reflectometry using a tunable laser. U.S. Patent 6,243,169 discloses an interferometric instrument for sensing the surfaces of a test object. U.S. Patent 6,252,669 discloses an interferometric instrument provided with an arrangement for periodically changing a light path of a received beam component. U.S. Patent 6,381,023 discloses an improved confocal microscope system which images sections of tissue utilizing heterodyne detection. U.S. Patent 6,423,956 discloses an angled-dual-axis confocal scanning microscope comprising a fiber-coupled, angled- dual-axis confocal scanning head and a vertical scanning unit. U.S. Patent 6,441,356 discloses an angled-dual-axis optical coherence scanning microscope comprising a fiber-coupled, high-speed angled-dual-axis confocal scanning head and a vertical scanning unit. U.S. Patent 6,445,939 discloses ultra-small optical probes comprising a single- mode optical fiber and a lens which has substantially the same diameter as the optical fiber. U.S. Patent 6,466,713 discloses the head of an optical fiber comprising the sensing probe of an optical heterodyne sensing device.
U.S. Patents 6,525,823; 6,559,950; 6,587,206; and 6,590,664 all disclose an optical system for monitoring a colloidal suspension.
U.S. Patent 5,094,534 discloses an optical layout similar to OCX, with the main puφose of measurement of displacement of a sensor diaphragm. However, no 3D image appears to be produced by the apparatus. U.S. Patent 5,202,745 discloses a reflectometer with the layout similar to OCX. In the invention, modulation of phase by PZX appears serving to produce signal spectral component of the same frequency as a fixed filter frequency in a detector, for any given mirror velocity. No 3D image is fonned.
U.S. Patent 5,268,738 discloses a reflectometer with multiple reference planes for increased depth range. U.S. Patent 5,268,741 discloses a reflectometer, bearing similarity to OCX. Modulation of the source is introduced for calibration. U.S. Patent 5,291,267 discloses a reflectometer, with similarity to OCX, comprising an amplifier for better sensitivity and reduced measurement times. U.S. Patent 5,365,335 discloses a reflectometer with similarity to OCX in the optical layout. U.S. Patent 5,459,570 discloses an optical coherence domain reflectometer employing two or more wavelengths of light. Polarization sensitivity provides for measurement of birefringence. Xhe apparatus employs modulation of light and a fixed filter. U.S.' Patent 5,719,673 discloses an apparatus for OCX characterization of a sample in a single spot. Xhe specific application is for measuring properties of different layers in the eye.
U.S. Patent 5,731,876 discloses a single spot OCX for measurement properties of layered films. U.S. Patent 5,835,642 discloses an optical delay line with all-fiber design. U.S. Patent 5,867,268 also discloses an optical delay line with all-fiber design. U.S. Patent 6,111,645 discloses an optical delay line. U.S. Patent 6,381,490 discloses an optical scanning and imaging system and related method for scanning and imaging an object. U.S. Patent 6,564,087 discloses a fiberoptic needle probe, which involves mechanical rotation. U.S. Patent 5,383,467 discloses a guidewire catheter and apparatus for diagnostic imaging. The apparatus is applicable to OCX and has a small probe diameter. U.S. Patent 5,570,182 discloses an OCX apparatus with acousto-optic or PZX modulation for heterodyning. It includes a movable lens for mechanical scan in the two dimensions normal to the optical axis. U.S. Patent 5,579,112 discloses a free space, non-fiberoptic, OCX apparatus. U.S. Patent 5,847,827 discloses a conventional OCX apparatus where the axial (z) coordinate of the focal spot follows the coherence scan. U.S. Patent 5,920,390 discloses an OCT apparatus with wavelength selectivity to characterize tissue based on the reflection spectrum. U.S. Patent 5,921,926 discloses an OCT apparatus with simultaneous spectral interferometry capabilities. The design involves mechanical rotation of the fiber/lenslet array. U.S. Patent 5,975,697 discloses an OCT apparatus with adjustable coherence properties and adjustable depth resolution. U.S. Patent 6,006,128 discloses an OCT apparatus with Doppler flow imaging. U.S. Patent 6,053,613 discloses a modified OCT interferometer. U.S. Patent 6,057,920 discloses an OCT apparatus with focal spot z-position following the
coherence scan. U.S. Patent 6,069,698 discloses an OCX apparatus with light path adjusted by the uniaxial stage such that the beam interference is detected for the scan range, to ensure stable acquisition of tomographic images. Xhe design involves mechanical motion of components.
U.S. Patent 6,124,930 discloses an improvement on conventional OCX. Xhe invention stabilizes frequency of the signal with a transverse scan. U.S. Patent 6,137,585 discloses an apparatus for generating data representative of a three-dimensional distribution of the light backscattering potential of a transparent or semi-transparent object such as a human eye. Xhe signal processing involves frequency shift by acousto-optical elements and heterodyning. One spot in the sample produces reference, another (moving) spot produces signal. Xhis design decreases sensitivity to eye movement. U.S. Patent 6,141,577 discloses an OCX apparatus with simultaneous spectral interferometry capabilities. Xhe design involves mechanical rotation of the fiber/lenslet array. U.S. Patent
6,175,669 discloses an optical coherence domain reflectometry guidewire with multiplexed fiber channels and both forward and side viewing. U.S. Patent 6,191,862 discloses a conventional OCX apparatus with improved highspeed axial (z) scanning. U.S. Patent 6,198,540 discloses an improved OCX apparatus with a plurality of reference planes, frequency-multiplexed, and high acquisition speed. U.S. Patents 6,201,608; 6,233,055; 6,252,666; and 6,307,633 all disclose an improved conventional OCX using a polarizing beam splitter and a polarization rotator to improve signal to noise ratio.
U.S. Patent 6,384,915 discloses multiplexed optical coherence reflectometry with multiple fibers for guidance and viewing. U.S. Patent 6,421,164 discloses a high speed conventional OCX apparatus with mechanical scanning. U.S. Patent 6,485,413 discloses an imaging system for performing forward scanning imaging for application to therapeutic and diagnostic devises used in medical procedures. U.S. Patent 6,501,551 discloses an imaging system for performing optical coherence tomography. Xhe endoscopic device involves mechanical rotation and longitudinal scanning. U.S. Patent 6,552,796 discloses a conventional OCX apparatus for reading information from a desired depth in a sample. U.S. Patent 6,552,797 discloses an apparatus for measurement of the freezing point of substances. U.S. Patent 6,564,089 discloses a conventional OCX apparatus with a Faraday rotator for stabilization of polarization and image contrast. U.S. Patent 6,608,684 discloses an OCX apparatus with magnetic mechanical lateral scanning of a fiber. U.S. Patent 6,608,717 discloses rapid in-vivo optical coherence microscopy with a conventional OCX layout. U.S. Patent 6,615,072 discloses a conventional OCX apparatus with a Faraday rotator for stabilization of polarization and image contrast. U.S. Patent 6,618,152 discloses an optical coherence tomography apparatus using optical- waveguide structure which reduces pulse width of low-coherence light frequency shift between reference and sample.
U.S. Patent 6,628,401 discloses an optical tomography imaging method and apparatus with amplified spontaneous emission light. Xhis is yet another implementation of conventional OCX with mechanical scanning. U.S. Patent 6,657,727 discloses interferometers for optical coherence domain reflectometry and optical coherence tomography using nonreciprocal optical elements. Xhe design uses differential signal processing. It involves mechanical scanning, as in conventional OCX. U.S. Patent 6,611,338 discloses an OCX apparatus using light of two wavelengths and amplitude modulation. Xhe design involves mechanical scanning. U.S. Patent 6,636,755 discloses a conventional OCX apparatus with high resolution for cellular imaging. Xhe design requires mechanical scanning. U.S. Patent 5,321,501 discloses a classic OCX apparatus by Fujimoto, et al. Xhe apparatus requires 2D mechanical motion of either sample or the probe (part of the probe). U.S. Patent 6,134,003 discloses an endoscopic OCX device, including a variety of probes. All embodiments involve mechanical motion.
U.S. Patent 5,465,147 discloses an OCT apparatus. The 3D image of the sample is produced by a coherence scan in one (z) dimension and 2D (x,y) conventional imaging with a CCD. A drawback of the apparatus is the potentially low-frequency detection typical of CCD.
U.S. Patent 5,994,690 discloses an OCT apparatus with deconvolution for improved axial resolution. The extent of resolution improvement was limited due to noise, as described in M. D. Kulkarni, C. W. Thomas, and J. A. Izatt, Image enhancement in optical coherence tomography using deconvolution, Electron. Lett., vol. 33, pp. 1365- 1367, 1997.
U.S. Patent ,6002,480 discloses OCT with deconvolution, similar to U.S. Patent 5,994,690.
U.S. Patent 6,295,132 discloses a reflectometer with acousto-optical deflectors and gratings for transverse displacement of the beam. The device is inherently large and is not fiberoptic.
U.S. Patent 5,555,087 discloses a 3D OCT apparatus employing a bundle of fibers, each producing the conventional OCT. The bundle replaces the mechanical 2D scan of other prior art such as in patents to Fujimoto and others. The drawbacks include large probe size, potentially slower operating with the CCD-type detector, and likely cross-talk between fibers producing artifacts in the output image. We now proceed to a review of optical coherence tomography.
The basic layout of optical coherence tomography is shown in Figure 1. A Light Source 110 has a low
coherence length Lc, related to the spectral width Δλ: L r « , where λ is central wavelength. Lc is typically ~10
Aλ μm when super-luminescent diode serves as the light source. Typical instrumental function of the OCT is presented in Figure 2. The instrumental function (impulse response) is defined as the signal produced by scanning (displacement) of a reflector 116 on a photo detector 124 of Figure 1 when the object 122 is a small point. The z-displacement 20 of the reflector 116 is scanned from the position when the optical path length from the light source 110 is exactly the same to the object 122 and to the reflector 116. The scanning range is typically comparable with the coherence length of the light source. An electromechanical device responsive to electromagnetic signals 118 such as but not limited to a piezo-electric stack is one possible means for causing reflector displacement 126, thereby introducing a phase delay. As the reflector is displaced, interference between the light scattered back from a point of object 122 and the light reflected by the reflector 116 produces z-fringes 142. As the displacement increases, the contrast of z- fringes 142 reduces due to loss of coherence (coherence gating). The width of the z-envelope 144 is defined by the coherence length Lc. Instrumental function, I (21) defines the resolution in z (depth) as the depth is scanned by displacement of the reflector 116. Lateral scanning of the sample is done mechanically, by displacing either the sample or the beam in the x,y directions. Illustrated also are first 114 and second 120 focusing elements, and a 2x2 optical coupler / splitter 112. Lateral resolution is defined by the spot size produced by second focusing element 120.
Conventional OCT and existing commercial products based on the OCT principles outlined above register the z-envelope 144 of the signal ("envelope-only" measurement). The resolution of the depth measurement (imaging) in these tools is comparable to the coherence length Lc.
Enhancement of the z-resolution can be achieved if phase of the signal is measured. An analogy may be made with rangefmders, which have two major classes: those based on time-of-flight of a short radiation pulse and those based on the measurement of phase shifts of long pulses of radiation. Displacement 126 of the reference reflector 116 of a standard OCT setup (FIG. 1) is related to time in this analogy. The OCT measurement resolution-
limited by the envelope of the signal is similar to the time-of-flight range measurement, whereas phase-shift measurements in range finding and phase demodulation in OCT have many similarities.
U.S. Patent 5,994,690 discloses an OCT system with signal processing, including phase demodulation, to increase the resolution of depth measurement. The system performs deconvolution of the measured signal by Fourier transform of the measured signal, division by the Fourier transform of the impulse response (instrumental function), and an inverse Fourier transform. The impulse response is obtained by replacing the sample with a reflector and measuring the response of the system. The drawback of the method is the possible effect of spectral variation of scattering by the obj ect. The result of deconvolution also strongly depends on the amount of noise and methods applied for its suppression. In experiments reported in M. D. Kulkarni, J. A. Izatt, Digital signal processing in optical coherence tomography, Coherence domain optical methods in biomedical science and clinical applications, Proc. SPIE; vol. 2981, pp. 2-6, 1997, an increase in resolution by a factor of slightly above 2 was achieved by deconvolution, whereas the number of carrier periods in the envelope was on the order of 10, the latter number being a reasonable expectation of gain in resolution with convolution in the absence of noise. OCT systems with improved signal to noise ratio (SNR) are of interest. U. Morgner, W. Drexler, F. X. Kartner, X. D. Li, C. Pitris, E. P. Ippen, J. G. Fujimoto, Spectroscopic
Optical Coherence Tomography, Optics Letters, 25, 111-113, January 15, 2000, presents a new method for enhanced resolution and extraction of color information using a Mortlet wavelet transform. Along with spectral measurement at every object point, 1 μm longitudinal resolution was achieved.
A typical OCT system known in the art combines ID coherence scanning and 2D mechanical scaiming to produce a 3D profile of the sample. It appears that throughout the prior art, interferometric scanning is performed in one dimension of the sample only, typically designated as the "z" depth dimension. Transverse scanning is performed by mechanical displacement of either the sample, or the beam over the sample surface. In large systems using conventional optics, mechanical beam scanning is typically performed by a deflector such as a galvanometer. In small, fiber optic OCT systems, mechanical rotation of miniature optical elements is used. Such a system is disclosed in U.S. Patent 6,445,939. Single-mode fibers and lenses with diameters substantially the same as the fibers are used in insertable medical devices, such as guidewires. Xhe need for mechanical motion of elements at the tip of the fiber probe adds complexity and cost to the device, and may compromise its range of applications and reliability.
Conventional OCX systems and existing commercial products based on the OCX principle register the envelope of the signal ("envelope-only" measurement). Xhe resolution of the depth measurement (imaging) in these tools is comparable to the coherence length Lc. As noted, for lateral x,y scanning today's "first generation" OCX instruments use mechanical systems to create cross-sectional images of the object. Xhese instruments have several limitations, including: Bulky probe tip • Slow operation
Limited depth of field Limited speed of image acquisition Rigid probe limiting the scope of applications
Large (mm) probe tip diameter, making the use of OCX problematic for many in-vivo imaging tasks • High cost of the OCX system
Low productivity and high cost of procedure
It would be desirable to provide OCX systems with 3D imaging through optical fibers by 3D coherence gating and scanning, with no mechanical motion at the fiber tip.
It would also be desirable to provide a fiberoptic OCX imaging system applicable to inserted devices such as endoscopes, bronchoscopes, needles, and other similar tools.
It would also be desirable to provide a fiberoptic OCX system with low-cost, removable and disposable fiber probe attachment.
It would also be desirable to provide a new OCX system with improved signal to noise ratio. . Such a system would have many advantages, including:
Xhree-dimensional, in vivo imaging of tissues and cells in real time at video frame rates • 3D coherence scanning with no probe mechanical motion
Focus-invariant imaging for extended depth of field 1 Improved, micron-level resolution for cellular imaging. All-fiber, low cross-section, flexible probes - "3D Camera Xhrough a Needle" Low-cost, disposable probes • Affordable, compact, lightweight system
Image acquisition with improved signal-to-noise ratio (SNR) to enable micron and sub-micron imaging resolution
Summary of the Invention Disclosed herein is optical coherence tomography with 3D coherence scanning, employing an apparatus with three or more fibers for object illumination and collection of backseattered light. Probe tips of the fibers are located in a plane normal to the optical axis. Light beams emerging from the fibers overlap at an object plane, intersection of the beams with the plane defining field of view of the OCX system. Interference of light emitted and collected by the fibers creates a 3D fringe pattern. Xhe 3D fringe pattern is scanned dynamically over the object by a combination of phase delays controlled remotely from near ends of the fibers opposite the probe tips of the fibers, and combined with light modulation,. Xhe dynamic fringe pattern is backseattered by the object, transmitted to the photo detector, and produces an AC signal on the output of the photo detector. Phase demodulation of the AC signal at selected frequencies and signal processing produce a measurement of a 3D profile of the object.
One embodiment of the invention is for a method and associated device comprising: a first optical fiber, comprising a first fiber tip thereof; a second optical fiber, comprising a second fiber tip thereof displaced from said first fiber tip substantially in a fiber tip plane substantially normal to an optical axis of said apparatus; and a phase delay φx module introducing a relative phase delay between said first optical fiber and said second optical fiber; wherein: adjustment of said phase delay φx module enables scanning of an object in a direction parallel to the displacement between said first fiber tip and said second fiber tip. Another embodiment of the invention with corresponding special technical features is for a method and associated device comprising: a first optical fiber, comprising a first fiber tip thereof; a second optical fiber, comprising a second fiber tip thereof displaced from said first fiber tip substantially in a fiber tip plane substantially normal to an optical axis of said apparatus; a third optical fiber, comprising a third fiber tip thereof displaced from said first fiber tip and said second fiber tip substantially in said fiber tip plane, such that said first fiber tip, second fiber tip and third fiber tip define said fiber tip plane; a first modulator for modulation of light passed through said first optical fiber; a second modulator for modulation of light passed through said second optical fiber; a third modulator for modulation of light passed through said third optical fiber further; and a non-linear light processing
system of at least of second power with respect to an intensity of light passed through said first modulator, said second modulator, and said third modulator, thereby producing a localized, 2-dimensional instrumental function.
A probe head with corresponding special technical features for use in connection with various embodiments of the invention is for a method and associated device comprising: a first optical fiber, comprising a first fiber tip thereof; a second optical fiber, comprising a second fiber tip thereof displaced from said first fiber tip substantially in a fiber tip plane substantially normal to an optical axis of said probe head; a third optical fiber, comprising a third fiber tip thereof displaced from said first fiber tip and said second fiber tip substantially in said fiber tip plane, such that said first fiber tip, second fiber tip and third fiber tip define said fiber tip plane; and an optical connector for connecting said probe head to and disconnecting said probe head from, said optical coherence tomography apparatus; wherein: said first fiber tip, said second fiber tip and said third fiber tip are configured such that light emitted therefrom substantially overlaps and is substantially unfocused at an intended distance of an object to be scanned from said first, second, and third fiber tips.
Brief Description of the Drawings Xhe features of the invention believed to be novel are set forth in the appended claims. Xhe invention, however, together with further objects and advantages thereof, may best be understood by reference to the following description taken in conjunction with the accompanying drawing(s) summarized below. Figure 1 schematically illustrates a typical layout of an OCX system according to prior art. Figure 2 is a depiction of a coherence-limited interference pattern defining impulse response of an OCX system.
Figure 3 is a schematic illustration of coherence transverse (x) scanning according to the present invention. Figure 4 is an illustration of a 3D interference pattern produced by two light sources. Figure 5 is an illustration of coherence gating in 3D with low-coherence light.
Figure 6 is a schematic layout of transverse (x) coherence gating with modulation and "running carrier." Figure 7 is a perspective view of a fiber tip layout for two-dimensional transverse (relative to z) coherence scanning in x and y directions.
Figure 8 illustrates a view of a low-coherence interference pattern produced by fibers in the layout of Figure 7.
Figure 9 is a perspective view of fiber tips layout for coherence scanning in x and y directions by three fibers, similar to Figure 7, and scattered light collection by a fourth fiber.
Figure 10 is a conceptual layout of a full 3 -dimensional OCX system in a preferred embodiment of the invention.
Figure 11 is a view of low-coherence interference patterns acquired by the configuration of Figure 10, at selected frequencies. Figure 12 is a view of the instrumental function produced by the configuration of Figure 10, suitably localized..
Figure 13 shows scanning of the system instrumental function of Figure 12 by controlled phase delay, in the x and y directions..
Figure 14 schematically illustrates the interchangeability of transmitting and receiving fibers in the various OCT system embodiments disclosed herein.
Figure 15 schematically illustrates a dual wavelength OCT system with improved signal to noise ratio. Figure 16 is a plot of a fringe width as a function of distance to the object.
Figure 17 schematically illustrates an embodiment comprising more than one low-coherence light source wherein the wavelengths of the sources are substantially similar.
Figure 18 schematically illustrates an embodiment comprising more than one low-coherence light source wherein the wavelengths of the sources are different. Figure 19 schematically illustrates an embodiment similar to that of Figure 18, wherein each source is modulated separately.
Figure 20 is a graph illustrating two dynamic fringe patterns resulting from the embodiment of Figure 19.
Figure 21 is a graph illustrating phase trajectories of the fringe patterns in Figure 20.
Figure 22 is an illustration showing how interference patterns shift under a suitable phase delay. Figure 23 is an illustration of a 3 -dimensional interference pattern, without localization of the instrumental function, further illustrating the component patterns which contribute to the overall pattern of Figure 8.
Figure 24 illustrates the non-linear (multiplicative rather than additive) composition used to achieve the localized instrumental function of Figure 12.
Figure 25 shown a full three-dimensional representation of the localized instrumental functions of Figures 12 and 24, with a phase delay in both the x and y dimensions.
Figure 26 illustrates a first embodiment of a probe head design for use in accordance with various embodiment of the invention.
Figure 27 illustrates a second embodiment of a probe head design for use in accordance with various embodiment of the invention. Figure 28 illustrates some modeling results of the OCT devices and methods disclosed herein, with deconvolution.
Detailed Description
Disclosed herein is a new fiberoptic OCT system with coherence 3D scanning of an object 122 to be scanned. The concept is based on using multiple fibers for illumination and/or collection of light from a sample. Referring to Figure 3, we begins by considering consider a first fiber 201 and a second fiber 202 of equal lengths connected to a common low-coherence light source 110, each emitting a diverging beam, so that there is an area where the beams overlap. In the area of overlap, interference of the beams will take place. A resulting x-fringe pattern 206 will vary in the x,y plane, normal to the regular OCT coherence scan in z-direction. Figure 4 illustrates the 3D x-fringe pattern produced by light from these two fibers, which can be made by considering cross-sections of the pattern by sides of a cube 220, with the light sources (fiber tips 1 and 2) at two corners 222 and 224 along the x axis. The scale of the fringes is arbitrary and is chosen for illustration puφoses only. For sources separated by a distance of ~ 100 wavelengths, the fringe pattern will have ~200 periods between the sources along the x-axis. The same interference pattern, with the effect of a finite coherence length light source, is shown in Figure
5. The contrast of the interference pattern reduces as the optical pathlength difference from the two sources at 222 and 224 to a point in the sample approaches the coherence length. This is similar to the effect of coherence length in the one-dimensional case, as in the prior art OCT systems illustrated in Figures 1 and 2.
With this type of configuration, coherence gating becomes possible, selective in the transverse direction, along the x axis. As shown in Figure 3, a phase delay ψλ module 210 introduced in fiber 201 and optionally a phase delay (p2 module 211 introduced in fiber 202 will shift the x-fringe pattern 206. Such a phase delay can be implemented by a variety of methods known in the art, including reflection from a movable reflector (116 and 118
in Figure 1) as commonly done in conventional OCT. Examples of devices suitable for introducing a phase delay include PZT-mounted reflectors, fiber stretching devices, grating based delay lines, polygonal scanners, rotary mirror arrays, and any similar or equivalent devices known or which may become known in the art.
Both the carrier signal (x-fringe pattern 206) and the x-envelope 208 will shift if a relative phase delay between the two fibers is introduced. Periodic (e.g., saw-tooth) variation of the phase delay will result in a periodic scanning of the fringe pattern. Similar to z-scanning in conventional OCT, the controlled phase delay illustrated in Figure 3 produces scanning in the x (or y) direction. The x-envelope 208 will scan as well, so that the relative phase of the x-fringe pattern 206 in the x-envelope 208 will stay unchanged. Intensity at a given point of the object 122 being scanned will experience modulation with the carrier frequency and varying amplitude (higher near the center of the x-envelope 208 and lower in the periphery).
Unlike in prior art, the transverse (x or y) coherence scanning can be performed by a remote control over the phases, at substantial distances from the fiber tips. This is important, because it eliminates the need for any mechanical scanning elements near the tips of the fibers. This makes the resulting probe device smaller, which is valuable for applications involving insertion into small areas, including endoscopy, colonoscopy, and optical biopsy. Additionally, a simple design of the fiber probe assembly makes possible its implementation as a low-cost, completely non-mechanical, removable probe head attachment 270 later illustrated in two embodiments in Figures 26 and 27, which may be disposable.
Figures 22 further illustrates how a phase delay introduced between the two light sources from fibers tips 1 and 2, as disclosed above, causes the interference pattern to shift. This should be contrasted with Figure 5 showing the same patterns in the absence of the phase shift.
A second important aspect of the invention introduces amplitude modulation of the intensities in the fibers, using first and second modulators 230, 232, as shown in Figure 6. For simplicity, Figure 6 illustrates the effects of such a modulation without the aforementioned phase delays. Later, phase delay and modulation will be combined into one system along with non-linear signal processing. Appropriate modulators include, but are not limited to, acousto-optical modulators (AOM). Other amplitude modulators, such as, but not limited to, electro-optical, semiconductor, magneto-optic, fiberoptic, and similar or equivalent modulators known in the art or which may become known in the art can also be used. Phase modulation can also be used instead of amplitude modulation. The types of modules described above for introducing phase delay can also be used to provide such phase modulation. Other phase modulators can also be employed, such as, but not limited to, photo-elastic, electro- optical, semiconductor, fiberoptic, and other similar or equivalent devices known or which may become Icnown in the art.
Optical spectrum of the modulated signal will contain, in addition to the optical frequency vopt , additional frequencies Vopt 4- fmod and V t — fmoά . Due to interference of optical signals with slightly different frequencies
(wavelengths), the interference pattern will contain components with moving fringes, so that intensity at a given point will oscillate with the frequency mod . The modulated light will have spectral components that create a moving fringe pattern 206, but the x-envelope 208 of the pattern will remain unchanged. That is, the phase delay discussed in connection with Figure 3 moves the envelope 208 but leaves the fringe pattern 206 unchanged within the envelope; while the modulation introduced in Figure 6 does not move the envelope 208 but does move the fringe pattern 206 with in the envelope. Deconvolution of the phase-demodulated signal produces the object profile in the x direction, similar to conventional OCT producing object profile in the z direction. The x and z profiles together provide for 2D coherence scanning.
Coherence scanning in both x and y directions can be produced if three fibers are used, resulting in a 3D fringe pattern illuminating the object 122. A possible layout is presented in Figure 7. A first fiber 201 (with first fiber tip 1) and a second fiber 202 (with second fiber tip 2) are employed, with tips oriented along the x axis produce scanning in the x direction. A third fiber 203 (with third fiber tip 3) is placed with its tip oriented along the y axis relative first fiber tip 1. The interference pattern produced by the first (201) and the third (203) fibers, scanned as discussed above, will produce a scan in the y direction. One or more of the fibers 1, 2, 3 (201, 202, 203) is also used for conventional OCT measurement to produce a z-scan of the object. Combining these x and y coherence scans with conventional z coherence scans then results in a fully 3D coherence scan, without mechanical motion near the fiber tips. Note, while the second and third fibers 202, 203 form an approximately 90 degree angle relative to the first and second fibers 201, 202 in Figure 7, that any configuration is possible so long as the tips of the three fibers define a fiber tip plane 71 substantially normal to the optical axis 72, i.e., so long as the three fiber tips do not all reside co-linearly along a single line normal to the optical axis 72. For example, not limitation, the three fibers might define the fiber tip plane 71 by forming an equilateral triangle, or an isosceles triangle, or, indeed, any sort of triangle. The details of the calculations to follow will then vary depending on the specific configuration chosen among the fiber tips, but in a way that will become apparent to someone of ordinary skill. The exemplary calculations to follow - and the general discussion to follow, for simplicity only - are based on the 90 degree angle (right triangle, with first fiber tip 1 opposite the hypotenuse) specifically illustrated in Figure 7, with the clear understanding that other configurations are possible so long as the three fiber tips define a plane. In general terms, Figure 7 thus illustrates a first optical fiber 201, comprising a first fiber tip 1 thereof; a second optical fiber 202, comprising a second fiber tip 2 thereof displaced from said first fiber tip 1 substantially in a fiber tip plane 71 substantially normal to an optical axis 72 of said apparatus; and a third optical fiber 203, comprising a third fiber tip 3 thereof displaced from said first fiber tip 1 and said second fiber tip 2 substantially in said fiber tip plane 71, such that said first fiber tip 1, second fiber tip 2 and third fiber tip 3 define said fiber tip plane 71.
Referring to Figure 7, interference patterns 208, 242, and 144 overlap in space to produce a 3D distribution of intensity. An x-y cross-section of this 3D fringe pattern 8 is shown in Figure 8, illustrating the interference patterns between respective fiber tip pairs 1-2 (x), 1-3 (y), and 2-3 (z). The linear composition of this pattern is further illustrated in Figure 23. Vertical fringes result from interference of light emitted by fiber tips 1 and 2 horizontal fringes result from fiber tips 1 and 3, and diagonal fringes result from fiber tips 2 and 3.
This may be further visualized by referring to Figure 23. With the addition of a third source 226, partially coherent with sources 224 and 222, the interference pattern becomes more complex. This pattern is now a sum of three interference patterns: Iu (x, y, z) , 723 (x, y, z) , and 713 (x, y, z) , produced by pairs of sources (224, 222),
(222, 226), and (224, 226), respectively. The component interference patterns 228 which result in the composite pattern 8 of Figures 8 and 23 are also illustrated. The pattern is not yet localized in space as would be desirable for imaging 3D scattering media by scanning the interference pattern in three dimensions. However, as will be shown later, additional signal processing can produce a localized instrumental function, and the phase delays discussed above can be used for non-mechanical scanning over the volume of the sample. Specifically, the signal processing will produce instrumental function equal to the product of component patterns I x > y, z) and I23 (x, y, z) .
Collection of light can be performed by one of the fibers 201, 202, or 203, or, alternatively, by a fourth, collection fiber 204 with a fourth fiber tip 4, as illustrated in Figure 9, which shows the detail of fiber tips near the sample (object) 122.
The conceptual layout of the overall 3D OCT system is shown in Figure 10, and is based on the four-fiber configuration of Figure 9 (although, again, a three fiber configuration is possible where at least one of the first three fibers 201, 202, 203 collects the light). Light from a wideband, low-coherence source 110 is sent into the first three fibers 201, 202, 203. Optical path lengths to the tips 1, 2, 3 of the three fibers are the same. The fourth, collection fiber 204 collects the light returning from object 122 (not shown in this illustration). The collected light is sent to a light processing system 108, which produces a signal proportional to the overall intensity of light scattered back by the object 122 and collected by fourth fiber 204 or one of the first three fibers 201, 202, 203. Each of the paths of fibers 202 and 203 comprise phase delay φx module 102 and phase delay φy module 104. Further, a phase delay φz module 106 is further introduced in a reference optical path 107 between source 110 and light processing system 108, and is used for conventional coherence scanning along the z axis as discussed in connection with Figure 1. First and second fibers 201, 202 also comprise first and second modulators 230 and 232 from Figure 6, while fiber 203 and reference optical path 107 comprise respective third and fourth modulators 234 and 236 as illustrated. Note that first fiber 201 does not need to have a phase delay module, though similarly to what was discussed in connection with Figure 2, it may optionally comprise such a phase delay module. What is important is that there be a relative phase delay between fibers. Note also, while phase delay φz module 106 and fourth modulator 236 are illustrated in reference optical path 107, that one or both of these can also be placed into an object optical path 109 carrying light between object 122 being scanned and light processing system 108. For z-scanning, object optical path 109 also joins light passing through reference optical path 107 with light striking and reflecting from the object 122 and passing through object optical path 109. In this way, the mixed optical path 111 carries a mixed reference and object signal. For z-scanning, a relative phase delay φz and a differential modulation is introduced as between reference optical path 107 and object optical path 109 before it light is delivered in a mixed 111 signal to light processing system 108.
In some embodiments to be discussed more fully later, the fiber tips 1, 2, 3, and (optional) 4 are assembled into a removable (and thereby disposable) probe head attachment 270, and attached via an optical connector 272 to the various elements shown to the left of optical connector 272 in Figure 10.
In the event that one employs the three-fiber configuration of Figure 7 and uses one of these three fibers for collection, the fourth fiber 204 is eliminated along with fourth fiber tip 4, and a three-fiber connection 260 illustrated by a dotted line in Figure 10 is used to split the return signal out of one of the three fibers 201, 202 or 203 (203 is used in the illustration), and to feed this return signal out toward light processing system 108. In all other aspects, the conceptual layout of Figure 10 remains the same. In either event, for z-scanning in addition to x,y scanning, there is an object optical path 109 joining light passing through reference optical path 107 with light striking and reflecting from obj ect 122.
The fringe pattern shown in Figure 8 is the impulse response, or instrumental function, /, 21 of the system. If the object 122 has only one, very small, scattering spot, the photo detector signal (assuming a linear detector for the moment) as a function of the spot x,y coordinate is proportional to the brightness of the fringe pattern of Figure 8. With real scattering objects, however, having scattering areas with large numbers of scattering spots, the signal reaching the photo detector is a convolution of the object density and the instrumental function. It is apparent from Figure 8 that the instrumental function is not localized around the center, which makes the process of deconvolution in signal processing challenging. A localized, narrow instrumental function is preferred.
Xhis problem can be solved by shifting frequency of the optical signal in fibers 201, 202, 203, and phase detection at combinations of the frequency shifts. Some frequency shift will take place due to introduced phase delay (210, 211 in Figure 3); the rate of change of the phase is the frequency added (or subtracted). In embodiments where the phase delay is introduced by a moving reflector similar to 116, 118, the frequency shift is the Doppler shift in the reflected light. Each optical frequency Vopt of the signal is transformed into vopt + fΩopp , where f)0„p is the Doppler shift. Additional frequency shift can be introduced by amplitude or phase modulation of the optical signals. In case of amplitude modulation at a frequency f , the spectrum of each original optical frequency is transformed into three spectral components: vopt + Dopp , vopt + Dopp + / , and Vopt + Dopp - / .
As noted earlier, besides similarities with the phase shift produced by phase delay, the difference is that amplitude-modulated light will have spectral components that create a moving fringe pattern, but the x- (or y-) envelope of the pattern will not move (contrast Figure 6 with Figure 3). Xhe phase of the "carrier" x- (or y-) fringe pattern will vary in relation to the x- (or y-) envelope with the frequency of modulation f . Delay introduced by a delay means (210 in Figure 3), on the other hand, results in a displacement of the envelope, with the phase of the fringe pattern relative to the envelope remaining constant. Light in fibers 201, 202, 203 and passing through fiber tips 1, 2, 3 is therefore modulated at different respective frequencies fλ , f2 , and f3 . Phase demodulation at the frequency fi °PP — fk °PP "*" (ft ~ fk)'ι ( ~ 1J2J3) will result in a different instrumental function for each demodulation frequency. Figure 11 shows the instrumental functions for each of the three detection frequencies: Figure 11, a) for demodulation at fu = **• - /2 Dopp + (/j - f2 ) ; Figure 11 , b) for fu = f?» - /3 Dopp + (f - f3 ) , and Figure 11, c) for f23 = f2 opp — f3 opp + (f2 — f3) . Xhe maximum values (brightness of the fringe pattern) are
Il2 = /j • I2 , Il3 = τjlλ ■ I3 , and I23 = ^jl2 ■ I3 , respectively.
Xhe instrumental function is still not localized around the center, which is desirable for deconvolution, as well as for envelope-only measurement. Xo produce the localized instrumental function, several processing options are possible. A first option is a light processing system 108 which multiplies, rather than adds, the signals detected at fl2 and f3 . In this approach, detection of the modulated optical signal produces two electrical signals, one detected at fn and the other detected at fi3 . Multiplication of the two electrical signals is then performed by one of the many approaches know in the art, for example, but not limited to, analog or digital multiplication. Xhe maximum intensity in the center of the envelope is in this case I12 • Il3 = Iλ • 72 • I3 . A second option is a light processing system 108 using a nonlinear photo detector, e.g., a quadratic photo detector with the output electrical signal proportional to (at least) the second power of the light intensity. While photodetectors are typically designed and optimized for high linearity, a separate broadband amplifier with nonlinear gain can easily be added to a linear detector for the desired second-harmonic detection. Both options will produce a signal with more than one spectral component, including V = fl2 + /13 = 2 i - f2 - f3 + 2 1 Dopp - 2 Dopp - 3 Dopp . Phase detection at this differential frequency produces a localized instrumental function 12, shown in Figure 12, which is much more desirable. Xhe central area
260, where brightness varies, is a 2D, x,y-envelope of the instrumental function. Intensity variation within the envelope is a product of the carrier fringe patterns in x and y directions (Figure 11, a, b).
Using a third option for light processing system 108, signal at the frequency vj1" can also be produced employing digital processing of the raw signals to accomplish the same results as in the above first and second options. In this case, sampling and analog-to-digital conversion of the modulated optical signal is produced at sampling rates substantially exceeding the fastest modulation frequency, and digital processing is used to produce the signal at the differential frequency, fn - fl3 .
Xhis instrumental function can be used in deconvolution as well as in envelope-only measurements. In both cases, modulation and phase shift near the light source in the optical path (Figure 10) produces non-mechanical 3D coherence scanning of the object.
Figure 24 illustrates the non-linear composition of the localized instrumental function 12 of Figure 12 by light processing system 108.
Referring back to Figure 10, transverse scanning in each direction (x or y) is achieved by introducing a phase shift in one of the fibers 201, 202, 203. Figure 13 x) shows scanning by a phase delay φx module 102 introduced in the optical path of fiber 202 (the fiber with tip 2 displaced along the x axis relative to tip 1). Figure 13 y) shows scanning by a phase delay φy module 104 introduced in the optical path of fiber 203 (the fiber with tip 3 displaced along the y axis relative to tip 1). Xhe introduction of these phase delays φx and φy is thus used to move the focal point of the scan of object 122 in the x and y directions, respectively. (Note again, that the 1-2 and 1-3 axes in all of these illustrations are orthogonal, but that any configuration where tips 1, 2, and 3 define a plane are acceptable.)
Xhis achieves an important advance over the prior art, by replacing the need for mechanical scans in the x,y plane with fully coherence-based scanning. Figure 25 shows a three-dimensional representation of the scan where a phase delay φx and a phase delay φy have both been introduced. Xhe third phase delay φz, module 106 is used to move the scan along the z axis in the conventional manner as described in connection with Figure 1, and the instrumental function is suitably-localized.
It is important to observe that this system as disclosed, in all embodiment, is fully reversible. Particularly, it should be noted that the optical paths providing illumination 110 of the sample and delivering backseattered light to the light processing system 108 are interchangeable. Xhus, for example, the two layouts shown in Figure 14 are equivalent and the functionality of the OCX system is exactly the same with either of these two layouts. (Only two fibers are shown in Figure 14 for simplicity of illustration.) One can interchange source 110 with light processing system 108 in Figure 10, thereby reversing all of the light transmission paths, and achieve exactly the same functional result. Note also that for any individual fiber, the phase delay module and the modulators can also be interchanged, i.e., that the order in which the light passing therethrough encounters these two elements does not matter. Xhe above analysis describes the proposed OCX system with "phase scanning" in the x and y dimensions of the object. Xhe required phase shift is achieved by combinations of phase delays φx , φy as shown in Figure 10.
Xhe amplitude modulations are introduced by modulators 230, 232, 234 including, but not limited to, acousto- optical modulators. An extension of the proposed method into 3D scanning of the object is achieved by introducing controlled phase shift (phase delay φz 106 and modulator 236 in Figure 10 - similar to the other three modulators), and processing the signals accordingly. Specifically, instead of multiplying a pair of signals as discussed earlier,
groups of three signals are multiplied. Alternatively, a non-linear photo detector with cubic (third power) non- linearity can be used. And, as earlier, digital processing of the raw signals may be employed to accomplish the same results achieved by a multiplier or a cubic photo detector.
Xhe shape of the z-envelope in the OCX signal (144, Figure 2) is important for measurement of the z- profile of the sample. In conventional OCX, the envelope is the instrumental function of the system defining resolution of the measurement. With phase detection, the envelope is a part of the system impulse response, used for deconvolution. If coherence length is large compared to the central wavelength, the envelope is wide, and the coordinate measurement becomes ambiguous since the whole number of fringes in the measured coordinate may be miscalculated, especially in the presence of noise. Short pulses with low coherence length are used in the art for high-resolution measurements. Xhe recent trend in OCX was to use shorter light pulses with shorter coherence length, such as Xi:Al203 lasers instead of super luminescent LEDs. Xhe new light sources produce short (fs) pulses, which makes it difficult to maintain the propagating pulse shape due to dispersion in the sample and in the media preceding it in the optical path.
Xhe alternative to short coherence lengths is a system that is capable of resolving the ambiguity of the absolute phase shift (including whole numbers of periods, or cycles) by other means. A similar problem in interferometry is solved in some cases by using light sources with more than one wavelength. In OCX, a similar approach can be implemented by using two detectors, each with a filter covering a part of the spectrum of the source. Xhe possible proposed layout of such a dual wavelength detection embodiment is presented in Figure 15. Once again, this is illustrated for the x-scan only, for simplicity. Xhe total spectral range covered by the two filters 152, 154 filtering wavelengths around different maxima
(central wavelengths) and an (optional) tap 156 is the same as in conventional OCX systems such as illustrated by Figure 1 and Figure 2. Parallel processing of the signals of photo detectors 153 and 155 (or more generally, the light processing system 108 if a multiplier or other signal processing is employed) will therefore produce the same resolution after correlation 157, deconvolution and processing 158. Xhe uncertainty of the number of whole
periods, however, can be reduced, with the effective wavelength λ
eff
Xhe width of each of the two components of the instrumental function, related to wavelengths λ{ , λ2 , λ2 Lc t « — — (z = l,2). Xhe filter bandwidth Aλt is yet another degree of freedom available for optimization. Both λf
Aλ[ are preferably substantially wide, comparable in bandwidth to the source, with overlapping passbands.
Signals for the two photo detectors measured simultaneously produce redundancy. Xhis may be used to reduce the effect of noise on the measurement, which greatly benefits the success of deconvolution and resolution of the 3D profile measurement of the object. In sum, light emerging from each of the at least two filters 152, 154 is measured substantially simultaneously and correlated, thereby providing an expanded spectral range for measurement, and redundant measurement with improved signal to noise ratio.
While two filters are shown here for illustration, more than two filters may be used, each sending a portion of the wavelength range of the backseattered light to a separate photo detector. A reference tap 156 and the photo detector 159 shown in Figure 15 are optional, to provide an additional measurement channel, serving as a reference to reduce the effect of any possible instability in the intensity of light entering light processing system 108.
In a related embodiment (given the reversibility of the system as discussed earlier), rather than (or in addition to) multiple filters, one can employ multiple light sources. Note that Figures 17, 18 and 19 to be discussed below are modeled on Figure 1 for simplicity, but can be employed in all of the embodiments disclosed herein.
Figure 17 illustrates an embodiment comprising more than one low-coherence light source 110-n, n=l, 2, . . . N to increase optical power and the related signal-to-noise ratio, frame rate, and other performance characteristics. Intensities of the sources are combined 171 to add incoherently, to produce one "efficient source" of the higher power equal to the power of the component sources.
Figure 18 illustrates an embodiment wherein the wavelengths of the sources are different, to produce a wider overall low-coherence spectrum. In this embodiment, the interference pattern remains the same, with maximum contrast at zero path length difference between the reference optical path 107 and the object optical path 109. Xhe coherence length, reciprocal of the spectral bandwidth, will be shorter, resulting in proportionally improved spatial resolution.
Figure 19 illustrates an embodiment wherein each source is modulated separately (modulators 191-n, n=l, 2, . . . N) to provide new possibilities in signal generation and processing. For example, modulating two sources at two close, but different frequencies creates two dynamic fringe patterns is illustrated in Figure 20. Detected synchronously, these two signals follow a phase trajectory as shown in Figure 21, determined by the modulation frequencies and coherence lengths of the sources, which can be use for suppression of noise, e.g., by correcting the 2D vector signal to the nearest point on the phase trajectory.
Y. Zhang, M. Sato, N. Xanno, Resolution improvement in optical coherence tomography by optimal synthesis of light-emitting diodes, Optics Letters, Vol. 26 Issue 4 Page 205, February 2001, noted earlier, provides some pertinent discussion of multiple LED devices in free space systems.
In some alternative embodiments, the transverse (x,y) scanning of a 3D fringe pattern to produce a 3D image of the sample (object 122) can be applied to other known types of OCX where presently mechanical scanning in the x,y plane is performed. One of such types of OCX is so-called "Spectral Radar" - an OCX sensor working in the Fourier domain (see, e.g., G. Hausler, M. W. Linder, "Coherence radar" and "spectra; radar" - new tools for dermatological diagnosis, J. Biomd. Opt. 3(1)21-31, 1998, and P. Andretzky, M. W. Linder, J. M. Hermann, A. Schultz, M. Konzog, F. Kiesewetter, G. Hauser, Optical coherence tomography by spectral radar: dynamic range estimation and in vivo measurements of skin, Proc. SPIE 3567, 78-87, 1998. Similar to the above, using at least three fibers for illumination and collection of light, to produce a 3D fringe pattern, combined with amplitude modulation and phase detection, will allow retrieving the 3D object profile instead of depth (z) profile measurement alone.
Xhe disclosure herein can also be applied in optical coherence microscopy (OCM). In OCM, unlike OCX, the sample may be opaque. Xhe 3D information about the surface scattering of the sample and therefore the 3D surface shape is characterized in OCM, whereas in OCX the 3D structure of the sample volume is imaged. Optionally, more than two fibers for each transverse scan direction (more than three total) can also be used, e.g., to improve the shape of the instrumental function of the apparatus. In particular, the effects of improved instrumental function and higher resolution may be provided by the operation of the multiple fiber sources as a optical phased array (OP A), as known in the art.
Depending on the size of the object and the distance to it, different modifications of the fiber tips 1, 2, 3 may be employed for object illumination at different distances. With no modification, the illuminated area of object 122 has the size 2NA • h , where NA is the fiber numerical aperture and h (see Figure 3) is the distance to the object. If larger areas or shorter distances are desired, beam expansion will be achieved by expanding the beam
using, for example, not limitation, lenses (spherical, aspheric, or gradient-index), tapered fibers, or similar or equivalent devices known or which may become known in the at.
When the distance h to the object is comparable or small compared to distance d between the fiber tips (see Figure 3), beam deflectors may be used to change the direction of light exiting from the fibers, so that the centers of the two light spots are both approximately coincident with the center of the object. Such deflectors may include, but are not limited to, prism attachments, fiber end polishing (or other preparation) at an angle, bending of the fibers, and reflecting surfaces. A common lens may also be used for all fibers, similar to dual- and multiple-fiber fiberoptic collimators.
Resolution of the OCX system depends on distance h to the object. When this distance is small compared to distance ^between the fiber tips, the period of the fringe pattern (Figure 3) is equal to half the wavelength. With signal processing and deconvolution, this may be an approximate value of the system resolution. For envelope-only measurement, resolution is worse, but still proportional to the fringe period, since a fixed number of fringes are contained in the envelope. Resolution is compromised as the distance to the object increases. Resolution can be improved by increasing the distance d between the fiber tips. Figure 16 shows calculated period of the fringe pattern ? as function of Λ, for the 1.3 micron wavelength and d= 125 microns. Xhese resolution estimates are for transverse (x,y) resolution. Resolution in depth (z) is the same as in conventional OCX systems described in the prior art.
Returning to Figure 10, it is again noted that the fiber tips are assembled into a removable (and thereby disposable) probe head attachment 270, and attached via an optical connector 272 to the various elements shown to the left of optical connector 272 in Figure 10. In this way, this new design provides for low-cost, disposable probe tips. Xhe probe needs no rotation or other mechanical motion, unlike existing OCX systems. Xhe probe is also flexible. Its cross-section can be very small, e.g., 0.25 by 0.25 mm if standard single-mode fiber (SMF) with 125 μm cladding diameter is used. Xhis translates into a 0.0625 mm2 cross-section, small enough for installing in needles and close to the cross-section of a standard 0.014" guidewire such as used in vascular diagnostic and interventional procedures. Since the core diameter of the standard SMF is only -10 μm, the 125 μm cladding diameter can be reduced even more, e.g., by chemical etching, to further minimize the probe size.
A detailed view and operation of an embodiment of fiber probe tip 270 suitable for use in accordance with various embodiment of the invention is shown in Figure 26. Xhe sample (object) 122 is illuminated by diverging light emitted by fiber tips 1, 2, and 3, which are beveled, deflecting the diverging beams and forming the 3D interference pattern substantially overlapping over a full field of view 266 at an intended distance of object 122 to be scanned from fiber tips 1, 2 and 3, as has been disclosed herein. Note that the light beams are unfocused when they strike object 266. Backseattered light (in a four-fiber system) is collected by the fourth fiber tip 4. (Recall that everything can also be reversed, see Figure 14) Each element of resolution ("3D pixel") in the sample (object 122) contributes to the backseattered optical signal with an amplitude defined by its reflectivity (scattering) on the one hand, and a phase defined by its location in relation to the illuminating fibers, on the other hand. Xhe 3D sample image is reconstructed from the detected signal by nonlinear light processing system 108 as discussed previously. Another possible implementation, as discussed earlier, is to use one (or more) of the fiber tips 1, 2, or 3 for collection of backseattered light, eliminating the for the fourth fiber tip 4.
Referring to Figure 26 (and Figure 27 to be discussed below), it should be noted that while illuminating fibers 201, 202, 203 may preferably be chosen single mode, the collection fiber 204 may preferably be multimode, thus providing higher light collection efficiency due to larger core diameter and higher numerical aperture.
An alternative embodiment of fiber probe tip 270 is shown in Figure 27. Xhe third single-mode fiber, displaced from the plane of the drawing, is present, similar to Figure 26, but hidden in the view presented. A probe head tip refractory element 262 changes the divergence and direction of the illuminating beams emerging from the single-mode fibers, and also improves collection of the backseattered light into the multimode receiver fiber. Figure 27 illustrates refractory element 262 inwardly refracting the diverging light from each fiber. While refractory element 262 in some circumstances may converge the light emerging from fiber tips 1, 2, and 3, it is important in all cases that this light substantially overlap over the full field of view 266 at an intended distance of object 122 to be scanned from fiber tips 1, 2, and 3, and that it not be focused at a single spot, i.e., pixel of sample 122. A length of an optional pellet 264 is selected to adjust the optimal distance to the tip of the receiver fiber, preserving the monolithic structure with no air gaps. All components are connected by standard means employed in fiber optics, such as thermal splicing or adhesion. Xhe lens focal length, its length, and the distance to the fibers provide several degrees of freedom for controlling illumination of the sample (angle between beams and related period of interference pattern) and collection of backseattered light. It is important to note again that the imaging remains unfocused, similar to the layout of Figure 26. Xhe depth of field limitation, typical of conventional OCX which requires focusing, is thereby not a factor. Xhe high resolution and image details in all three dimensions result instead from digital processing of the photodetector signals.
It should be noted again that the optical paths providing illumination of the sample (object) 122 and the path delivering backseattered light to the photodetector are interchangeable, as discussed in connection with Figure 14. When the system is reversed, fourth fiber tip 4 serves as the sole source of illumination of the sample (object) 122, and fiber tips 1, 2, and 3 collect backseattered light to produce interference on the surface of the photodetector, similar to conventional OCX. "Effective interference pattern" is created in the sample (object) 122 in both conventional OCX and the OCX devices and method disclosed herein. Xhe difference is that the devices and methods disclosed herein produce a 3D scan with localized instrumental function and without physical movement, as opposed to ID coherence scanning of conventional OCX. Xhe amount of light on the photodetector in the 3D OCX disclosed herein with focus-free imaging may be lower, compared to standard OCX with a lens that focuses light in the sample (object) 122. However, with the smaller volumes under examination typical of cellular imaging, the amount of light is quite sufficient. A radiometry calculation performed in preliminary studies is described below related to signal deconvolution.
Resolution of the proposed OCX devices and method disclosed herein depends on the distance to the object. As the distance to the point-in-sample increases, imaging resolution will deteriorate, for two main reasons: decreasing signal-to-noise ratio and increasing fringe width. Xhe effect of signal-to-noise ratio is discussed in detail below. Xhe fringe width increases with the distance h as 0.5/I /1 + h l 2d) , where λ is the central wavelength and dis the distance between illuminating fiber tips. For values typical of cellular imaging, h is comparable to d, and expansion of the fringes is not dramatic. Resolution can be improved by increasing the distance d between the fiber tips of the probe. Xhe tradeoff between resolution and probe is one of the parameters optimized in system design.
Image acquisition rate is limited by the electronics and processing only, which is superior to the present OCX systems which largely rely on mechanical scanning of the focal spot. By maximizing the acquisition rate, real-time video rates at 30 frames per second should be feasible. Coherence scanning in all three dimensions of the proposed 3D OCX, unavailable to the existing OCX technologies, will enable a breakthrough improvement in acquisition speed.
Xhe disclosure is also applicable to elimination of mechanical scanning in X and Y dimensions in existing OCX systems with no scanning in the depth dimension, such as "spectral radar." (P. Andretzky, M. W. Lindner, W. Michael, J. M. Herrmann, A. Schultz, M. Konzog, M. F. Kiesewetter, G. Hausler, Optical coherence tomography by spectral radar: dynamic range estimation and in-vivo measurements of skin, Proc. SPIE 3567, p. 78-87, 1999. Image acquisition speed of the disclosed system should exceed that of the present state of the art by at least an order of magnitude due to phase scanning replacing the conventional mechanical X,Y scanning.
Deconvolution in OCX, described in the literature, produced limited improvement in the image resolution, due to noise. A factor of 2.5 increase in "image shaφness" was obtained at a loss of dynamic range of 2 dB using iterative restoration algorithm. (A. M. Rollins, M. D. Kulkarni, S. Yazdanfar, R. Ungarunyawee, J. A. Izatt, In vivo video rate optical coherence tomography, Optics Express, Vol. 3, No. 6, pp.219-229, 1998; and M. D. Kulkarni, C. W. Xhomas, J. A. Izatt, Image enhancement in optical coherence tomography using deconvolution, Electronics Letters, vol. 33, No. 16, pp. 1365-1367, 1997.) Xhe recent progress in ultra low noise photodetectors, driven largely by fiberoptic telecommunications applications, is promising to enable more efficient deconvolution in OCX. In addition, the approach to cellular imaging with deconvolution disclosed herein differs from attempts described in the literature by reducing the field of view and depth of imaging, which is justified by the gain in resolution, reaching the micron length scale.
An important limitation of conventional OCX systems is due to the use of a focusing lens. In such conventional systems, certain lens design issues, such as aberrations affecting focusing and coupling of backseattered light into receiver fiber, typically define practical distances between the probe and the sample in the millimeter range. As a result, the signal-to-noise ratios (SNR) are typically relatively low. Xhis compromises the efficiency of image deconvolution, as was just discussed.
In the present invention, light in the sample is not focused, and due to the sample illumination over the over a full field of view 266 (see Figure 26) rather than on a single focal spot, and processing of backseattered light with an unfocused, 3D fringe pattern, closer distances are possible between the probe and the sample. As a result, higher SNR becomes possible, enabling efficient image deconvolution. Xhis ultimately results in higher resolution, potentially at the micron and submicron levels, as illustrated in the modeling results presented in Figure 28 as discussed below. Xhis capability is of particular value for cellular imaging and inspection of microscopic structures Figure 28 presents preliminary modeling suggesting that micron-level resolution is likely to be feasible, to enable cellular imaging in 3D OCX. Xhe calculations assumed an SLD light source with the 1300 nm wavelength, 40 nm spectral bandwidth (coherence length Lc=42 μm), and 30 mW optical power. Numerical aperture of the single-mode fibers is 0.1, diameter of the receiver fiber 100 μm, object feature is square 5 μm on the side, resolved to 1 μm; reference optical power is 30 μW. Xhe photodetector was assumed Analog Modules 712A-4, with sensitivity 77 V/μW and noise 20 fw/Hz'/2. Refractive index of different element of cells varies from 1.38 in rat liver cells (J. Beuthan, O. Minet, J. Helfman, G. Muller, The spatial variation of the refractive index in biological cells, Physics in Medicine and Biology, vol. 41, pp. 369-382,, 1996) to 1.7 in melanin. (I. Vitkin, J. Woolsey, B. Wilson, and R. Anderson, Optical and thermal characterization of natural (sepia oficinalis) melanin, Photochemistry and Photobiology, vol. 59, pp. 455-462, 1994.) Xhe calculation assumed Δn=0.05 between the feature being imaged and the background, resulting in intensity reflection of approximately 0.1%; frame rate 30 FPS, scattering in the tissue 1 mm"1 (S. J. Matcher, M. Cope, D. X. Delpy, In vivo measurements of the wavelength dependence of tissue-scattering coefficients between 760 and 900 nm measured with time-resolved spectroscopy
Applied Optics, Vol. 36, No. 1, pp. 386-396, 1997), the distance to the feature inside the sample varied between 0.2 and 1 mm. Xhe target resolution was 1 μm, which defined the amount of optical power from a "pixel" and the
signal to noise ratio. Xhe detected signal intensity was calculated accounting for losses due to geometry, scattering, modulation, and polarization effects.
Retroreflection by the object (0.1%) was assumed to be a step function 5 μm wide, sampled at the 1 μm resolution; the OCX impulse response (or point spread function, PSF) was sampled similarly. Convolution of the reflection and PSF was produced by multiplying their Fast Fourier transforms (FFX). Next, bandwidth-limited noise was added to the convolution. Detection bandwidth was selected equal to the Nyquist frequency. Next, deconvolution of the noisy signal was performed by dividing the Fourier transform of the noisy OCX signal by the Fourier transform of the PSF and applying the inverse FFX. Xhe results for thee sample distances are presented in Figure 28. At a distance of 0.5 mm, imaging cells with 1 μm resolution and small effect of noise appears possible.
Xhe resolution increase due to proper filtering and deconvolution is encouraging, considering the assumed coherence length Lc= 42μ m. Xhe 3D field of view of 90x90x90 results in nearly megapixel 3D image sizes in each frame; with the assumed frame rate of 30 FPS, the video stream will be approximately 30 MB per second. Also shown is the modeling for 0.2 mm, which shows an even smaller noise effect. At a distance of 1 mm, the SNR value of 85 is not sufficient for effective deconvolution, and the noise effect is detrimental.
Xhe above results illustrate a mathematical model available for designing the 3D OCX setup. Xhe model can be employed for selection of components with a clear understanding of their effect on the overall system perfonnance. Requirements to components may be specified using the model. Xhe effect of wavelength, spectral range, and power of the SLD can be immediately estimated. Xradeoffs between performance characteristics such as depth range, frame rate, resolution (feature size and number of pixels), and others can be selected using the model. Several other technical considerations are also recognized, as follows:
Speckle effects are known to be a problem in standard OCX. Speckles appear when unresolved features of an object scatter coherent light with significant mutual phase shifts. Xhe phase shifts vary as the observation angel changes, resulting in apparently random intensity modulation of the image. In the high-resolution 3D OCX proposed herein, the resolved feature size will be comparable to the wavelength. Xherefore, the effect of speckles can be expected to be small.
Regarding fiber dispersion, for efficient light collection and improved signal-to-noise, the receiver fiber (204 in the four-fiber embodiments) preferably is multimode, with the core diameter larger than in single-mode fibers. Modal dispersion, or differential mode delay (DMD) is the result of different rays of light traveling over different pathways within the fiber core. Since each pathway has a different length, the time of arrival of each ray is a function of the distance traveled. DMD in multimode fibers is typically on the order of 1 ps/m (J. E. George, S. Golowich, P. F. Kolesar, A. J. Ritger, and M. Yang, Laser Optimized Multimode Fibers for Short Reach 10 Gbps Systems, National Fiber Optic Engineers Conference, 2001 Xechnical Proceedings, pp. 351-361). Modulation frequencies in 3D OCX should not exceed 100 MHz. Xypical length on the fiber is ~lm. Xhe resulting conservative estimate of the DMD-related phase shift in our case is therefore 10"4. Xhis amount is small compared to 2%, and the effect of DMD on the system performance should be negligible.
It is expected that other dispersion effects, such as polarization mode dispersion (PMD) in the fiber stretcher, will be negligible as well due to relatively low (tens of MHz at most) modulation rates.
Note that while at least three fibers are required for 3 -dimensional OCX scanning, the disclosure herein can also be applied to 2-dimensional scanning, in the x-z or x-y planes. Xhis is useful for situations in which a two- dimensional scan is all that is required, for example, in scanning thin films, or in taking two-dimensional cross-
sectional scans of a three-dimensional object where motion is supplied along the third dimension, such as a material being extruded.
All-fiber imaging as disclosed herein with dramatically reduced cross-section will enable the first "camera through a needle". Xhe reduced size and weight of the 3D OCX and flexible probe with low cross-section will allow for in-vivo histological imaging of tissues and cells, in a minimally invasive procedure. Focus-invariant registration of low-coherence interference patterns will provide extended depth of field unavailable with present
OCX.
Xhe technical issues of existing OCX result in a reduced scope of its commercial applications, currently limited to ophthalmology. Xhe 3D OCX disclosed here with simple, passive, flexible, low-cost, and disposable probes will enable practical applications for functional biological imagery (three-dimensional imagery of the sample on a cellular scale) and commercial applications, including clinical biomedical imaging for diagnostics of cancer and pre-cancer and cardiovascular diagnostic imaging.
Acquisition and processing at video frame rates will enable real-time 3D OCX imaging and storage of large numbers of cross-sectional views of living cells. Xhis will enable real-time monitoring of cell dynamics in biological and biophysical experiments. In medical applications, the 3D OCX combined with optional spectroscopic imaging will provide real-time "virtual biopsy" to be presented to the user with head-mounted stereoscopic display or in traditional multiple cross-sectional views as in CX-scan. .
Resolution of existing OCX systems employs "envelope only" measurement with the resolution defined by the coherence length of the source, typically on the order of 10 microns. Due to noise, prior attempts to apply deconvolution to OCX signals did not result in dramatic resolution enhancement. Xhe proposed development of high-resolution 3 OCX, fully applicable to cellular imaging, will now be based on a novel probe geometry and leveraged by the recent progress in low-noise optoelectronics and new commercially available components.
Xhis disclosure can help to create probes using diffused light-scattering principles that can monitor the eye, skin, and brain health of living beings. Xhis can aid research work centered on the eventual design of a goggle-like device that combines the dynamic light scattering probe with other techniques including corneal autofluorescence, laser Doppler velocimetry, reflectometry and oximetry, optical coherence tomography, and laser scanning technology (http://www.medicaldesignnews.com/full_story .php?WID=652). Xhis technology will also serve as a tool of functional imagery with enhanced capability to image functioning biological systems at the cellular length scale, with three-dimensional imagery of the sample. Finally, 3D OCX will be applicable for diagnostics of the performance of labs-on-a-chip, including detecting the presence of bubbles and particles and removing or characterizing them, as well as measurement of fluidic movement.
Biomedical imaging for diagnostics and minimally invasive surgery is a significant application. Xhe market of diagnostic biomedical imaging grows at a steady rate of over 10 percent annually and is expected to reach
$20 Billion in 2007. OCX offers resolution of imaging by at least order of magnitude better than ultrasound, X- Rays, CAX scan, or MRI. Xraditional OCX has found very limited commercial applications, largely due to the mechanical motion in the probe. Known commercial OCX system, produced by Zeiss-Humphrey Systems, appear to be limited to diagnostics of the eye retina. It appears that no commercial OCX tools for other parts of the body currently exist, leaving over 90 percent of the potential commercial applications unaddressed. With the flexible, low-cost, disposable probes, the disclosed 3D OCX will enable the first OCX tool for commercial application in medicine, beyond ophthalmology. Xhe probe cross-section compatible with standard guide wire will enable applications in cardiovascular diagnostic imaging. In oncology, Pap Smear test alone is nearly a billon dollar market. Xhe "3D OCX camera through a needle" with optional spectroscopic capability will produce "biopsy
movies" to be presented to a radiologist in the familiar microscopic format, as well as an optional "virtual reality" real-time imaging with a head-mounted display. In minimally invasive surgery, the 3D OCX can work in combination with conventional endoscopes to provide "close-up" three-dimensional views of cells in the areas of interest. Fiber probes can also be attached to surgical instruments (e.g., scissors or lancets) for imaging of tissues and cells ("real-time biopsy") during surgery.
Other potential biomedical applications are in ophthalmology; developmental biology; dermatology; dentistry; gynecology; urology; gastroenterology; laryngology; surgical guidance and intervention. Non-biomedical applications of 3D OCX are envisioned in high-density data storage, polymer matrix composites, inspection systems for semiconductor and MEMS manufacturing, and industrial inspection systems. Another application is real time 3D imaging of plant cells. A plant cell is much more mobile than an animal cell. Movement within a cell may be deep within tissue.
Microgravity effect on the immune system are also of interest. For example, extended duration spaceflight produces reduced immune response, possibly through alteration of basic cellular functions. Results indicate that by day four of a spaceflight mission, the cell-mediated immune system is noticeably degraded. Changes in cell shape are not monitored at the early stages, so it is unknown when such changes take place. Research in cellular biology, particularly in the area of cells that experience microgravity, indicate that changes in cells that adapt to microgravity could be initiated by physical changes that occur in the first few seconds/minutes of that environment. Furthermore, increased sensitivity to bacterial allergens is a particular problem. Studies have not revealed whether this is due to stress or microgravity. Xhe proposed development will permit the studies of the effects of microgravity on cell samples in the earliest stages of exposure to microgravity, providing information for fundamental space biology and bioastronautics.
3D OCX can also be applied for examining biological structures in biochips. Plastic molded biochips show great promise as high-throughput screening and point-of-care bio-sensing technologies. Xoday these on-chip chemical and biochemical analysis schemes are limited by the availability of highly sensitive detection methodologies. 3D OCX with several multiplexed probes will enable speedy and robust examination of biochips. In addition, Doppler velocimetry, available in 3D OCX as well as in traditional OCX, will provide high-resolution noninvasive velocity determinations. 3D OCX will facilitate detection in volumes ranging from 10s of microliters to 10s of picoliters, thus reducing sample volume requirements and enabling a broad range of materials to be analyzed. Other applications requiring real-time 3D imaging of living cells will be accommodated by the proposed technology as well.
While only certain preferred features of the invention have been illustrated and described, many modifications, changes and substitutions will occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.