One or more embodiments of the present invention relate generally to optical imaging, in particular, methods of frequency-domain optical coherence tomography (OCT) in which the interference signal is measured as a function of optical frequency. The invention is a novel method for suppressing artifacts in OCT images, and in blurring the mirror image present in images acquired by frequency-domain OCT so that it can be clearly distinguished from the real image.
Optical Coherence Tomography (OCT) is a technology for performing high-resolution cross sectional imaging that can provide images of tissue structure on the micron scale in situ and in real time. OCT is a method of interferometry that uses light containing a range of optical frequencies to determine the scattering profile of a sample. The axial resolution of OCT is inversely proportional to the span of optical frequencies used.
In recent years, it has been demonstrated that frequency domain OCT has significant advantages in speed and signal to noise ratio as compared to time domain OCT (Leitgeb, R. A., et al., Optics Express 11:889-894; de Boer, J. F. et al., Optics Letters 28: 2067-2069; Choma, M. A., and M. V. Sarunic, Optics Express 11: 2183-2189).
In frequency domain OCT, a light source capable of emitting a range of optical frequencies excites an interferometer, the interferometer combines the light returned from a sample with the light returned from a reference reflector, and the intensity of the combined light is recorded as a function of optical frequency to form an interference spectrum. A Fourier transform of the interference spectrum provides the reflectance distribution along the depth within the sample. Frequency domain OCT requires some means to record the interference spectrum, the intensity of light output from the interferometer as a function of optical frequency. Current methods of Frequency domain OCT can be divided into two categories.
In spectral-domain OCT, a grating or prism or other means is used to disperse the output of the interferometer into its optical frequency components. The intensities of these separated components are measured using an array of optical detectors, each detector receiving an optical frequency or a fractional range of optical frequencies. The set of measurements from these optical detectors forms an interference spectrum (Smith, L. M. and C. C. Dobson, Applied Optics 28: 3339-3342). Typically the light source emits a broad range of optical frequencies simultaneously.
Alternatively, in swept-source OCT, the interference spectrum is recorded by using a source with adjustable optical frequency, with the optical frequency of the source swept through a range of optical frequencies, and recording the interfered light intensity as a function of time during the sweep (Swanson, E, et al., U.S. Pat. No. 5,321,501).
Frequency-domain OCT efficiently uses the light returned from a range of depths within the sample, as all the returned light contributes to the modulation in the interference spectrum. This is in contrast to the case of time domain OCT, in which the interference signal is sensitive only to light returned from depths in the sample that match the current length of the reference arm to within the coherence length of the source. For a given illumination level and time of exposure, the achievable signal to noise ratio is substantially greater in frequency-domain OCT than in time-domain OCT (Leitgeb, R. A. et al., Optics Express 11: 889-894; de Boer, J. F. et al., Optics Letters 28: 2067-2069; Choma, M. A., and M. V. Sarunic, Optics Express 11: 2183-2189).
Chromatic dispersion is a property of an optical element that characterizes the degree by which the optical path length through that element varies across a range of optical frequencies. OCT determines the position in the sample of a scattering center based on the difference in optical group delay between two optical paths: 1) the path of light scattered from the sample, and 2) a reference optical path. Most OCT literature refers simply to the difference between sample and reference optical path lengths; but when one considers chromatic dispersion one must distinguish between the phase delay and the group delay associated with a given optical path length. OCT is sensitive to the difference in group delay (see, for example, section 2.1 in Fercher et al., Optics Express 9: 610-615). OCT necessarily uses a range of optical frequencies. If the chromatic dispersion is not matched between the two paths, the apparent position of the scattering center depends on the optical frequency used. A mismatch in chromatic dispersion thus broadens the axial resolution of the OCT as explained by Hitzenberger et al., (Journal of Biomedical Optics 4: 144-151). For this reason, in most OCT systems the chromatic dispersion is closely matched between sample and reference paths (see, for example, U.S. Pat. Nos. 6,385,358, 6,615,072, 6,618,152) sometimes through the use of dispersive optical devices (see, for example U.S. Pat. No. 6,282,011). One of the advantages of quantum OCT (a method of time-domain OCT described by Teich et al. in U.S. Pat. No. 6,882,431) is that it partially cancels the mismatch in chromatic dispersion between the sample and reference paths.
Since a perfect match of chromatic dispersion is not simple and can add cost to a frequency domain OCT system, numerical correction of the residual mismatch in chromatic dispersion has been described by Fercher et al. (Optics Express 9: 610-615; Optics Communications 204: 67-74) and by Marks et al. (Applied Optics 42: 204-217). The preceding authors used a test sample to determine the residual mismatch in chromatic dispersion. Alternatively, the numerical correction for dispersion can be empirically adjusted for best sharpness of the resulting OCT image, without using a test sample, as described by Wojtkowski et al. (Optics Express 12: 2404-2422).
Frequency domain methods of OCT use the fact that interference between light scattered from the sample and the reference beam causes spectral interference fringes, a modulation in the intensity of the combined beam as a function of optical frequency. The spacing of the interference fringes depends on the difference in optical group delay between the light scattered from the sample, and reference light. In addition to chromatic dispersion match or correction, there are two additional major issues associated with this method.
The first one is that the spacing of the interference fringes does not depend on which of the two paths is longer. Therefore, simple methods of frequency-domain OCT do not distinguish between scattering from a sample location corresponding to an optical group delay a certain amount longer than that of the reference path, and scattering with an optical group delay the same amount shorter than that of the reference path. The resulting image contains the true scattering profile plus the superposed mirror image of the scattering profile. (See, for example, Yun et al., Optics Express 12:4822-4828.)
Phase shifting one of the interfering beams has been proposed to remove the mirror image in spectral domain OCT, but to date these methods leave some residual image and are complicated and costly (Leitgeb, R. A. et al., Optics Letters 28: 2201-2203; Izatt et al. U.S. Patent Publication No. 2004/0239938; co-pending U.S. patent application Ser. No. 10/933,795; Sarunic, et al., SPIE 5316: 241-247). In particular, sequential phase shift does not produce a good result if the sample such as a human eye tends to move during the period when the multiple phase interference spectrums are recorded (Targowski et al., Optics Communications, 229:79-84). On the other hand, simultaneous parallel multi-phase detection of the interference spectrum would require the use of at least two spectrometers or two detector arrays in the spectrometer, which will not only add complication in terms of optical alignment but also substantially increase the spectrometer cost. Frequency shifting one of the interfering beams in swept-source OCT has been used to shift the image away from its mirror image (Yun et al., Optics Express 12: 4822-4828), but this method is applicable only in situations where the sample being imaged has finite visible extent, and in which the detection electronics can accept the resulting higher-frequency signal. In addition, it requires an additional optical frequency shifter which is an expensive item and hence would add cost to the OCT system.
The second issue is that when the interference spectrum is recorded as a function of optical frequency, periodic interfering signals as a function of optical frequency can corrupt the digitized spectrum. For example, in the case of spectral-domain OCT, the readout electronics of the array detector often adds periodic pattern noise to the recorded optical intensities. In swept-source OCT, electronic clocks and counters can add a similar periodic error signal. Furthermore, a spectral ripple of the light source, such as those caused by etalon effects from inadvertent reflections in path from the light source, causes a periodic error signal in the recorded interference spectrum.
The periodic error signal from these and similar sources is difficult to distinguish from the interference fringes caused by light scattered from the sample. If present in the spectra when the image is reconstructed, the periodic error signal will cause the artificial appearance of a scattering center at a particular depth in the reconstructed image. The periodic error signal can be partially removed by measuring it separately from the measurement on the sample, and subtracting them from measurements on the sample. Such correction methods work only to the extent that the periodic error signal repeats exactly, in both amplitude and phase, on each spectrum acquisition. Signals from non-synchronized clocks, for example, will not repeat exactly and cannot be cancelled in this way.
From the above discussions, it can be seen that previous frequency-domain OCT systems generally require a match in the chromatic dispersion characteristics of the two interferometer paths and even with a perfect match or a correction of the residual mismatch, the existing methods cannot cost-effectively address the issue of a mirror image and the artifacts induced by periodic spectral noise from either the source or the detector, there is hence a need in the art of frequency domain OCT to remove the stringent requirement of chromatic dispersion match, to easily and clearly distinguish the mirror image from the real image and to substantially suppress the image artifacts due to periodic spectral error signals.
One or more embodiments of the present invention satisfy one or more of the above-identified needs in the art. In particular, one embodiment of the present invention is a method for optical coherence tomography wherein the chromatic dispersion differs significantly between sample path and reference path, and the measured interference spectrum is multiplied by a complex phase factor to compensate for the chromatic dispersion.
One aspect of the present invention is to use the blurring effects of chromatic dispersion on OCT images to allow the viewer to distinguish the real image from the mirror image artifact in the resulting OCT tomogram, these blurring effects being corrected by multiplication by the complex phase factor only for the true, non-mirrored image.
Another aspect of the present invention is to use the blurring effects of chromatic dispersion on OCT images to substantially suppress the artifacts in the real image due to periodic interfering signals in the frequency-domain OCT measurement
Another aspect of the present invention is the use of different materials in the sample and reference arms of the OCT system to achieve a difference in chromatic dispersion between these arms.
Another aspect of the present invention is to use the blurring effects of chromatic dispersion on undesired artifacts in OCT images, so as to provide a measure of the degree of rejection of these artifacts.
These and other features and advantages of the present invention will become more readily apparent to those skilled in the art upon review of the following detailed description of the preferred embodiments taken in conjunction with the accompanying drawings.
Implementations of swept-source OCT are described by Swanson and Chinn (U.S. Pat. No. 5,956,355) and a modern implementation of spectral-domain OCT is described by Wojtkowski, et al. (Optics Express 12: 2404-2422). The common features of these two frequency-domain OCT techniques are reviewed here.
In frequency-domain OCT, the interference is recorded as a function of optical frequency. In spectral-domain OCT, detector 130 separates the combined light into its optical frequency components, using a spectrometer and a camera or other means, and provides as output the interference spectrum as a function of optical frequency. In swept-source OCT, source 101 is swept through a range of optical frequencies, and the output of detector 130 at each time during a sweep is associated with the optical frequency concurrently output from source 101.
Referring to
For wavelengths near 800 nm, the optical path length in glass fiber varies with optical wavelength by approximately 0.12 fs/nm per mm of fiber (Saleh, B. E. A, and M. C. Teich Fundamentals of Photonics, chapter 5). Multiplication by the speed of light gives this dispersion in terms of optical path length in vacuum: 0.036 μm/nm per mm of fiber. Therefore the 100 mm extra fiber causes the optical path length difference, between sample and reference arms, to vary across the spectrum by 3.6 μm/nm. Considering that light passes the extra fiber twice between leaving the source and reaching the detector, the complete optical path length difference, between the complete sample and reference paths, varies by 7.2 μm/nm. A broadband source with full-with at half-maximum intensity equal to 30 nm gives 10-micron axial resolution in OCT (using equation (1) in Fercher, et al., Optics Communications 204: 67-74). The mismatched dispersion we consider in this example causes the apparent location of a scattering center to vary by 100 μm across the 30 nm bandwidth of the source.
Referring again to
Alternative to the Michelson interferometer architecture shown in
In
Let the light incident on the sample have electric field amplitude E0. Then the light scattered from the sample and accepted back into the interferometer has electric field
where r(z) is the amplitude-reflection coefficient of the sample at depth z, q=2πn/λ is the spatial optical frequency of the light in the sample, the variable λ denotes the wavelength of the light in vacuum, n is the average refractive index within the region of interest in the sample, and nS is a function along the path S with values equal to the refractive indices of the various materials along the sample path.
The equation in the preceding paragraph divides the phase accumulated by the light wave into two parts. The integral containing r(z) accounts for the phase accumulated from the end of path S to the individual scattering centers in the sample. The integral over the path S accounts for the phase accumulated within the instrument. The path S is along the optical path from the source to the reference depth z=0 in the sample, and then from the depth z=0 in the sample following the optical path to one of the detectors. The refractive index nS is a function of position determined by the materials along the optical path S.
In general the refractive index in the sample n depends on both position z and on λ, but in the preferred embodiment we define q=2πn/λ using a spatially averaged refractive index n that depends on wavelength λ only. Inclusion of the wavelength-dependent but spatially averaged refractive index in the definition of q compensates for the spatially averaged chromatic dispersion of the material in the sample. Marks et al. (Applied Optics 42: 204-217, section 3) describe reconstruction methods to correct for non-uniform chromatic dispersion within the sample, which methods can optionally be used in conjunction with the present invention.
Similarly, the light returned from the reference path has electric field
where α is a real number relating the amplitude of the light returned by the reference path to that incident on the sample, the path of integration R is along the optical path of the reference light from source to the detector, and the refractive index nR is a function of position determined by the materials along the optical path R.
The two integrals over paths R and S add a net optical phase, denoted by φD, in addition to the phase determined by the structure of the sample through r(z):
The phase mismatch φD is a function of optical frequency q. For ease of understanding the method, it is convenient to take the paths R and S to have the same group delay, meaning that dφD/dq=0 at the center of the spectrum used for OCT. The important quality of φD is its higher-order derivatives in q.
The optical power from one port of the interferometer, at optical frequency q, is
where |ER|2 is the power entering the spectrometer from the reference arm, and |ES|2 is returned from the sample, out of |E0|2 power incident on the sample. The final term noise represents measurement noise and interfering signals. Interfering signals from system clocks, for example, can be introduced into the record of P(q) before digitization, causing artifacts that are easily visible after Fourier transform. The power P(q) is measured by the light detector as a function of optical frequency, and digitized for processing. The first term, the reference arm power, has the same spectral shape as the source, so it can be cancelled by subtracting a spectrum measured with the sample arm blocked. The last term is the power returned from the sample, and is negligibly small compared with the other terms, given appropriate choice of reference arm power |ER|2. In cases where |ES|2 is not small, it is suppressed by the numerical compensation of dispersion, described below, in the same way as the term denoted by noise. The two middle terms are the interference terms; it is difficult to isolate only one of these terms without making a separate measurement with a different phase shift φD between sample and reference arm. In frequency-domain OCT systems that do not use phase shifting, the two interference terms produce two separate images of the sample. The alternate configuration shown in
The derivative of phase with respect to frequency is generally called group delay. Herein we use the term group delay to denote the derivative of the phase φ with respect to spatial optical frequency q. Thus, the equation above says that the apparent depth varies across the spectrum by one-half the variation in group delay.
a illustrates the spectrum P(q) that would be measured at one output of an interferometer with a single scattering center in the sample path and an intentionally introduced dispersion mismatch of φD=100 μm2q2+20 μm3q3. The range of optical frequencies q shown in
The phase appears in the expression for P(q) above only in the argument of the cosine function. The cosine function is not sensitive to the sign of its argument, so the spectrum P(q) could equally well have come from a scattering center at depth D≡−(dφ/dq)/2. The dashed curve in
After the interference terms are isolated, they are multiplied by a factor e−iφ
Other numerical methods for correcting the effects of dispersion mismatch have been described elsewhere. The method presented by Wojtkowski et al. (Optics Express 12: 2404-2422) includes the additional step of a Hilbert transform before multiplication by a correcting phase factor. The method presented by Fercher et al. (Optics Communications 204: 67-74) performs a convolution on the spatial-domain signal r(z) instead of the equivalent multiplication on the spectrum P(q).
Numerical Fourier transform of the corrected interference spectrum yields the desired coefficient of reflection r(z), plus two undesired images from the other terms. The phase φD due to mismatched dispersion varies across optical frequencies q, which causes these undesired images to be blurred. The term noise has a contribution from readout electronics of the detector or detectors, and this contribution often has a periodic structure, which produces sharp peaks at constant z in the image if not dealt with otherwise; the phase factor introduced numerically desirably blurs these sharp peaks, reducing their peak intensity.
a illustrates apparent depth of the scatter considered in
The axial resolution of an OCT system is typically defined as the full-width in z at half-maximum of the magnitude of the coefficient of reflection |r(z)| determined by the OCT measurement, with a sharp reflector as the sample. The bandwidth of the source is one limit to the axial resolution; for a Guassian source with full-width at half maximum intensity Δλ in wavelength, the axial resolution δz is
(Wojtkowski et al., Optics Express 12: Page: 2404-2422). The corresponding relation in terms of optical frequency q is
The range of optical frequencies recorded by the detector also limits the axial resolution. Following the convention of Wojtkowski et al. (Optics Express 12: Page: 2404-2422) we require that reflections separated by half the axial resolution, δz/2, produce interference spectra that differ by at least one fringe in the number of fringes produced within the range of optical frequencies recorded. Under this convention the axial resolution is limited by
where ΔΛ denotes the range of wavelengths recorded and ΔQ the corresponding range of optical frequencies. We define the axial resolution of an OCT system to be the minimum δz supported by both the source width and the range of recorded optical frequencies.
The image artifacts caused by periodic interfering signals are reduced in intensity by the correction of mismatched chromatic dispersion. Consider a particular periodic signal,
noise=A cos(fq)×e−iφ
which would cause a spike of amplitude A in the Fourier transform if there were no correction factor e−iφ
Applying the correction factor e−iφ
The apparent depth D of the scattering centers varies by approximately 450 μm over the spectral range of ΔQ≈0.7/μm that was recorded to produce the image in
The degree of mismatch in chromatic dispersion determines the rate at which the relative optical group delay varies across the spectrum, so an OCT system that records spectra over a larger spectral range ΔQ will show a greater total variation in relative group delay for a given amount of dispersion mismatch. Alternatively, an OCT system that records spectra over a larger spectral range ΔQ requires less chromatic dispersion mismatch to produce the same total variation in relative group delay.
In images such as that shown in
a illustrates a situation in which the tissue has been moved down in the field of view, and in which the dispersion mismatch has not been corrected. The horizontal artifact 603 at the center of the image is the residual artifact 503 described above. The second horizontal artifact 604 comes from periodic noise in the detection electronics.
b illustrates the results of numerical compensation. The desired image is sharpened, while the mirror image is further blurred. The second horizontal artifact 604 is blurred and reduced in intensity.
Chromatic dispersion mismatch can also be used to compensate for non-uniform sampling of the interference spectra. In spectral-domain OCT the combined light is typically dispersed by a prism or diffraction grating and measured by a camera. The dispersion mechanism maps optical frequency onto position on the camera, but this mapping is in general non-linear and the samples of the interference spectrum P(q) are non-uniform in q. In swept-source OCT optical frequency varies as a function of time, and the mapping from optical frequency on to the time of measurement is in general non-linear. Interference spectra P(q) are conveniently sampled uniformly in time, which sampling will be non-uniform in q.
Typically the interference spectra are re-mapped from the coordinate along which they were measured onto a quantity linear in q. The resampling operation is computationally expensive, with operation count comparable to the Fourier transform. Resampling involves interpolation between measured data points, which can be performed using Fourier transform methods, or by interpolation kernels as described by Smith and Gossett. Wojtkowski et al. (Optics Express 12: 2404-2422) disclose a related technique in which the resampling step, normally used for correction of non-uniform sampling, is adjusted to compensate for residual dispersion mismatch. Instead, one can choose the mismatch between dispersion in the paths of the interferometer so that the variation in frequency of the fringes compensates for the non-uniformity of sampling. By way of example, a spectrometer in approximately Littrow configuration covering the frequency range from 750 nm to 850 nm will encode optical frequency q onto position x approximately as x=q−0.15 μm q2. If such a spectrometer is used with an interferometer having a phase mismatch φD=360 μm2 q2 due to chromatic dispersion, then the recorded fringes have the form
cos(2qz+φD)≈cos[2xz+(360 μm2+0.30 μm z)q2].
At the position z=−1.2 mm, the spectrometer mapping cancels the effect of mismatched dispersion, giving fringes uniformly spaced as a function of x, so a direct Fourier transform of the data sampled in x will provide an image that is sharp near z=−1.2 mm. This method uses less processing to produce an image that is sharp over a limited range of depths. Specifically, the image remains sharp so long as the coefficient (360 μm2+0.30 μm z) is less than the square of the axial resolution of the OCT system, as explained by Fercher et al. (Optics Communications 204: 67-74). For a system with 10 μm resolution, the image remains sharp on this 10 μm scale over a range of depths z=−1.2 mm±0.3 mm. An analogous method is available in swept-source OCT, where the potentially nonlinear relation between instantaneous optical frequency of the source and time takes the place of the nonlinear mapping between optical frequency and position along the spectrum.
Another application of chromatic dispersion mismatch is in conjunction with phase shifting methods to remove the mirror image. Phase shifting methods to remove the mirror image are described, for example, by Leitgeb. et al. (Optics Letters 28: 2201-2203). As mentioned earlier, one difficulty of sequential phase shift methods is that motion of the sample causes an un-controlled phase shift between exposures. An exemplary method of phase shifting is to measure the same tissue with OCT twice, with the phase of the reference beam changed by nominally 90° between measurements. Apparatus to perform the phase shifting are described in co-pending U.S. patent application Ser. No. 10/933,795. We denote the two measurements of the interference spectra by
A(q)=∫r(z)ei2qzdz+∫r(z)e−i2qz-i2φ
B(q)=β∫r(z)ei2qzdz+β*∫r(−z)ei2qz-i2φ
where β is the complex-valued ratio of the electric field in the reference beam for the first measured phase to the electric field for the second measured phase, and β* is its complex conjugate. Often the ratio β has a phase that is proportional to optical frequency q (β=ρe−i2qΔ with constants ρ and Δ) because typical methods of producing the phase shifts by changing the length of the reference arm will cause a phase shift proportional to q, as will a displacements of the sample.
Given the two measurements A and B one forms the linear combination using the best available estimate b of the actual ratio β
in which the unwanted mirror image is suppressed to the extent that b≈β. Errors as small as 1% in the estimation of β are typically visible on conventional OCT images, because these images typically contain useful information spanning several orders of magnitude in signal intensity; a mirror image only 1% the amplitude of the true image will still be within the dynamic range typically presented, and visible in the logarithmic grayscale typically used for OCT image presentation. One particular cause of an error in b is motion of the sample between recording measurement A and recording measurement B.
Published work has attempted to estimate the true ratio β by adjusting the parameter b so as to optimize the image reconstructed from combination C (for example, Cyganek et al., Proceedings of SPIE 5316:248). These methods require a metric to judge the correctness of the resulting image. When chromatic dispersion is matched, the mirror image produces a symmetric image, and image asymmetry has been used as a metric for the degree of removal of the mirror image. Highly asymmetric images are then judged more likely to be correct.
When the chromatic dispersion is significantly mismatched, the mirror image is blurred compared to the desired image, and thus the mirror image has lower contrast. Measures of image contrast can be used to judge the degree to which the mirror image has been removed.
The measure of contrast used to generate
∫|r(z)|4dz/(∫|r(z)|2dz)2,
where the reflectivity profile r(z) was determined from the linear combination C(q). Alternate measures of contrast can be used. Other operations made directly on the longitudinal scan can be used, such as the peak intensity or an entropy measure such as ∫|r(z)|2log|r(z)|dz. Alternatively, if the image is expressed as an array of pixels, statistics on the distribution of pixel intensities, such as the variance, skew, or kurtosis, can be used as measures of contrast.
Although various embodiments that incorporate the teachings of the present invention have been shown and described in detail herein, those skilled in the art can readily devise many other varied embodiments that still incorporate these teachings. For example, the difference in chromatic dispersion, effected by use of differing optical materials in the sample and reference paths in the preferred embodiment, can alternatively be achieved using any optical element that provides significant chromatic dispersion; the Gires-Tournois interferometer, and the pair of diffraction gratings, are two devices often used for this purpose (Siegman, A. E., Lasers, section 9.3).
The following references are hereby incorporated herein by reference.
US Patent Documents
U.S. Pat. No. 6,385,358 Everett et al. Birefringence insensitive optical coherence domain reflectometry system
U.S. Pat. No. 6,615,072 Izatt et al. Optical Imaging Device
U.S. Pat. No. 6,882,431 Teich et al. Quantum Optical Coherence Tomography
U.S. Pat. No. 6,618,152 Toida Optical coherence tomography apparatus using optical-waveguide structure which reduces pulse width of low-coherence light
U.S. Pat. No. 6,282,011 Tearney, et al. Grating based phase control optical delay line
U.S. Pat. No. 5,321,501 Swanson, E., et al. “Method and apparatus for optical imaging with means for controlling the longitudinal range of the sample”
U.S. Pat. No. 5,956,355 Swanson, E., and S. R. Chinn, “Method and apparatus for performing optical measurements using a rapidly frequency-tuned laser”
U.S. patent application Ser. No. 10/933,795 Everett, M, et al., “A patterned spinning disk based optical phase shifter for spectral domain optical coherence tomography”
U.S. Patent Publication No. 2004/0239938 Izatt et al., “System for Fourier domain optical coherence tomography”
Other Publications
Choma, M. A., M. V. Sarunic, C. Yang, and J. A. Izatt (2003). “Sensitivity advantage of swept source and Fourier domain optical coherence tomography.” Optics Express 11(18): 2183-2189.
Marta Cyganek, M., M. Wojtkowski, P. Targowski, and A. Kowalczyk (2004) “Numerical estimation of the total phase shift in Complex Spectral OCT in vivo imaging” Coherence Domain Optical Methods and Optical Coherence Tomography in Biomedicine VIII, Proceedings of SPIE 5316:248-251.
De-Boer, J. F., B. Cense, B. H. Park, M. C. Pierce. G. J. Tearney, and B. Bouma, (2003). “Improved signal-to-noise ratio in spectral-domain compared with time-domain optical coherence tomography.” Optics Letters 28(21): 2067-2069.
Fercher A. F, C. K. Hitzenberger, M. Sticker, R. Zawadzki, B. Karamata, T. Lasser (2001). “Numerical dispersion compensation for Partial Coherence Interferometry and Optical Coherence Tomography.” Optics Express 9: 610-615.
Fercher, A. F., C. K. Hitzenberger, M. Sticker, R. Zawadzki, B. Karamata, T. Lasser (2002). “Dispersion compensation for optical coherence tomography depth-scan signals by a numerical technique” Optics Communications 204: 67-74.
Hitzenberger, C. K., A. Baumgartner, W. Drexler, A. F. Fercher, (1999). “Dispersion Effects in Partial Coherence Interferometry”, Journal of Biomedical Optics 4(1): 144-151.
Leitgeb, R. A., et al. (2003). “Performance of Fourier domain vs. time domain optical coherence tomography.” Optics Express 11(8): 889-894.
Leitgeb, R. A., et al. (2003). “Phase-shifting algorithm to achieve high-speed long-depth-range probing by frequency-domain optical coherence tomography.” Optics Letters 28(22): 2201-2203
Marks, D. L., A. L. Oldenburg, J. J. Reynolds, S. A. Boppart (2003). “Digital algorithm for dispersion correction in optical coherence tomography for homogeneous and stratified media” Applied Optics 42: 204-217.
Saleh, B. E. A, and M. C. Teich, (1991) Fundamentals of Photonics, John Wiley and Sons.
Siegman, A. E., Lasers, University Science Books, Sausalito Calif., 1986.
Smith, J. O., P. Gossett (1984) “A Flexible Sampling-Rate Conversion Method” Proceedings (2): 19.4.1-19.4.4, IEEE Conference on Acoustics, Speech, and Signal Processing, San Diego, March 1984.
Smith, L. M. and C. C. Dobson (1989). “Absolute displacement measurements using modulation of the spectrum of white light in a Michelson interferometer.” Applied Optics 28(15): 3339-3342
Targowski, P., M. Wojtkowski, Z. Kowalczyk, T. Bajraszewski, M. Szkolmowski, I. Gorczyńska, (2004) “Complex spectral OCT in human eye imaging in vivo” Optics Communications 229:79-84.
Wojtkowski, M., T. Bajraszewski, P. Targowski, A. Kowalczyk (2003). “Real-time in vivo imaging by high-speed spectral optical coherence tomography.” Optics Letters 28(19): 1745-1747.
Wojtkowski, M., V. Srinivasan, T. H. Ho, J. G. Fujimoto, A. Kowalczyk, J. S. Duker (2004). “Ultrahigh-resolution, high-speed, Fourier domain optical coherence tomography and methods for dispersion compensation” Optics Express 12: Page: 2404-2422.
Yun, S. H., G. J. Tearney, J. F. de Boer, B. E. Bouma, (2004) “Removing the depth-degeneracy in optical frequency domain imaging with frequency shifting” Optics Express 12:4822-4828.
This application claims the benefit of the filing date under 35 U.S.C. § 119(e) of Provisional U.S. Patent Application Ser. No. 60/645,791, filed on Jan. 21, 2005, and to Provisional U.S. Patent Application Ser. No. 60/714,721, filed on Sep. 7, 2005, which are hereby incorporated by reference in their entirety.
Number | Name | Date | Kind |
---|---|---|---|
5321501 | Swanson et al. | Jun 1994 | A |
5956355 | Swanson et al. | Sep 1999 | A |
6282011 | Tearney et al. | Aug 2001 | B1 |
6385358 | Everett et al. | May 2002 | B1 |
6615072 | Izatt et al. | Sep 2003 | B1 |
6618152 | Toida | Sep 2003 | B2 |
6882431 | Teich et al. | Apr 2005 | B2 |
7102756 | Izatt et al. | Sep 2006 | B2 |
7256894 | Chen et al. | Aug 2007 | B2 |
20030137669 | Rollins et al. | Jul 2003 | A1 |
20040239938 | Izatt | Dec 2004 | A1 |
20050171438 | Chen et al. | Aug 2005 | A1 |
20060192969 | Marks et al. | Aug 2006 | A1 |
Number | Date | Country | |
---|---|---|---|
20060171503 A1 | Aug 2006 | US |
Number | Date | Country | |
---|---|---|---|
60714721 | Sep 2005 | US | |
60645791 | Jan 2005 | US |