U.S. patent application number 13/723006 was filed with the patent office on 2016-09-08 for system and method for imaging subsurface of specimen.
The applicant listed for this patent is Manish Kulkarni. Invention is credited to Manish Kulkarni.
Application Number | 20160256051 13/723006 |
Document ID | / |
Family ID | 50975426 |
Filed Date | 2016-09-08 |
United States Patent
Application |
20160256051 |
Kind Code |
A9 |
Kulkarni; Manish |
September 8, 2016 |
SYSTEM AND METHOD FOR IMAGING SUBSURFACE OF SPECIMEN
Abstract
There is a need for robust and portable system, apparatus and
method for imaging subsurface of specimens. We have described a
modular OCDR-OCT system and OFDR-OCT system to obtain high quality
images. The instant application also discusses proprietary
algorithms that have been modified from existing algorithms and
their use as a combination to suit a particular system. The imaging
of stationary, moving and combination of both subsurface structures
such as retina for diabetic patients is described.
Inventors: |
Kulkarni; Manish;
(Pleasanton, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Kulkarni; Manish |
Pleasanton |
CA |
US |
|
|
Prior
Publication: |
|
Document Identifier |
Publication Date |
|
US 20140180075 A1 |
June 26, 2014 |
|
|
Family ID: |
50975426 |
Appl. No.: |
13/723006 |
Filed: |
December 20, 2012 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12732484 |
Mar 26, 2010 |
|
|
|
13723006 |
|
|
|
|
12706717 |
Feb 17, 2010 |
|
|
|
12732484 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61B 5/42 20130101; A61B
3/12 20130101; A61B 3/1241 20130101; A61B 5/7225 20130101; A61B
5/02007 20130101; A61B 1/24 20130101; A61B 3/102 20130101; A61B
5/004 20130101; A61B 5/0066 20130101; A61B 5/08 20130101; A61B
5/0088 20130101; A61B 5/7221 20130101 |
International
Class: |
A61B 3/12 20060101
A61B003/12; A61B 5/02 20060101 A61B005/02; A61B 1/24 20060101
A61B001/24; A61B 5/00 20060101 A61B005/00; A61B 5/08 20060101
A61B005/08 |
Claims
1. A system, comprising: a light source emitting light of a
specific bandwidth called a first light; the first light is sent to
a specimen using a source arm and sample arm; a beam splitter to
spilt the first light from the source arm as a first path light to
a reference arm and as a second path light to the sample arm; a
fiber optically integrated birefringent reference mirror returning
the first path light to the beam splitter to join a returning light
from the specimen; an optical delivery unit and a waveplate to
receive the second path of light at the sample arm; the sample arm
sends the second path of light to the specimen and the specimen
reflects back the second path of light as a returning light via the
optical delivery unit and the waveplate to the beam splitter; a
partial returning light from the beam splitter travels through a
detector arm to a volume-phase holographic grating unit and a
detector array; the volume-phase holographic grating unit disperses
the partial returning light from the beam splitter and a dispersed
light enters the detector array to produce a light spectrum; and a
processor to perform a data analysis using a specific algorithm on
the light spectrum to form an image of the specimen.
2. The system of claim 1, further comprising: a .lamda./8 waveplate
located on the sample arm.
3. The system of claim 1, wherein the fiber optically integrated
birefringent reference mirror is at least one of .lamda./8
fractional wave mirror, and .lamda./8 Faraday rotator mirror.
4. The system of claim 1, wherein a specimen is at least one of
retina, skin, anterior segment of the eye, gastrointestinal tract,
lungs, teeth, blood vessels, subsurface area of semi-conductors,
chip manufacturing and sensitive medical equipment.
5. The system of claim 1, wherein the specific algorithm is at
least one of frequency resampling, demodulation, dispersion
compensation, and Doppler processing, wherein dispersion
compensation comprises of coherent deconvolution, wherein frequency
resampling step comprises of convolution using a Kaiser-Bessel
window, wherein demodulation step comprises of a modified Hilbert
transform.
6. The system of claim 1, further comprising: the optical delivery
unit in the sample arm creates scan patterns, wherein the
scan-pattern comprises of at least two B-scans, each B-scan having
its specific A-scan rate.
7. The system of claim 5, wherein Doppler processing step includes
short time Fourier transforms computation in lateral direction.
8. The system of claim 5, wherein Doppler shift is estimated by
computing centroid of the short time Fourier transform spectrum
using power near the spectral peak, which is an adaptive centroid
algorithm.
9. The system of claim 5, wherein the Doppler processing step
estimates blood flow velocities for augmenting diagnosis of
diabetic retinopathy.
10. A system, comprising: a tunable light source producing a light
of various frequencies within a specific bandwidth called a first
light; the first light is sent to a specimen using a source arm and
sample arm; a beam splitter to spilt the first light from the
source arm as a first path light to a reference arm and as a second
path light to the sample arm; a mirror returning the second path
light to the beam splitter to join a returning light from the
specimen; an optical delivery unit and a waveplate to receive the
second path of light at the sample arm and send it to the specimen
and the specimen reflects back the second path of light as a
returning light via the optical delivery unit and the waveplate to
the beam splitter; a partial returning light from the beam splitter
travels through the detector arm to a detector; the detector to
convert the partial returning light from the beam splitter into an
electric current; an analog to digital convertor to digitize the
electric current into a digitized electric current; and a processor
to perform a data analysis using a specific algorithm on a
digitized electric current to form an image of the specimen,
wherein the specific algorithm is at least one of resampling,
demodulation, dispersion compensation, Doppler processing and
inverse Fourier transform.
11. The system of claim 10, further comprises: a fiber optic
.lamda./8 waveplate attached to the rear end of the fiber in the
sample arm.
12. The system of claim 10, wherein a specimen is at least one of
retina, skin, anterior segment of the eye, gastrointestinal tract,
lungs, teeth, blood vessels, subsurface area of semi-conductors,
chip manufacturing and sensitive medical equipment.
13. The system of claim 10, wherein the fiber optically integrated
birefringent mirror is at least one of fractional wave mirror,
mirror, free space mirror and Faraday rotator mirror.
14. The system of claim 10, wherein Doppler processing algorithm
includes Short time Fourier transforms computation in lateral
direction.
15. The system of claim 10, further comprising: the retina scanned
by performing concentric circles at a variety of speed.
16. A method, comprising: sending a specific bandwidth light from a
light source to a specimen using a source arm, and sample arm;
splitting the specific-bandwidth-light using a beam splitter from
the source arm as a first path light to a reference arm and as a
second path light to the sample arm; returning the second path
light from a fiber optically integrated birefringent mirror to the
beam splitter to join a returning light from the specimen;
receiving the second path light at the sample arm using an optical
delivery unit and a waveplate and sending the second path light to
the specimen; reflecting back the second path light from the
specimen as a returning light via the optical delivery unit and the
waveplate to the beam splitter; transporting a partial returning
light from the beam splitter through the detector arm to a
volume-phase holographic grating unit and a detector array;
dispersing the partial returning light from the beam splitter using
the volume phase holographic grating unit to produce a dispersed
light and entering the detector array to produce a light spectrum;
and performing a data analysis using a specific algorithm including
at least one of frequency resampling, demodulation, dispersion
compensation and Doppler processing on the light spectrum to form
an image of the specimen using a processor.
17. The method of claim 16, wherein a specimen is at least one of
retina, skin, anterior segment of the eye, gastrointestinal tract,
lungs, teeth, blood vessels, subsurface area of semi-conductors,
chip manufacturing and sensitive medical equipment.
18. The method of claim 16, wherein the fiber optically integrated
birefringent mirror is at least one of fractional wave mirror,
mirror, free space mirror and Faraday rotator mirror.
19. The method of claim 16, further comprising: performing a data
analysis using the dispersion compensation algorithm includes the
process of coherent deconvolution.
20. The method of claim 16, further comprising: scanning the retina
by performing concentric circles at a variety of speed.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] The instant application is a continuation-in-part
application and claims priority to pending U.S. patent application
Ser. No. 12/732,484, filed on 26 Mar., 2010 and U.S. patent
application Ser. No. 12/706,717, filed on 17 Feb. 2010. The
disclosure is hereby incorporated by this reference in its entirety
for all of its teachings.
FIELD OF TECHNOLOGY
[0002] The following description relates to a system, method and an
apparatus for imaging and evaluating the microstructure of any
specimen on the subsurface. More specifically the description is
relevant to imaging of biological specimen such as a retina in
diabetic patients using optical coherence domain reflectometry
(OCDR), optical frequency domain reflectometry (OFDR), optical
coherence tomography (OCT), Doppler processing and Doppler OCT
technology in combination.
BACKGROUND
[0003] Optical Coherence Domain Reflectometry (OCDR) has been
playing a major role in industrial, scientific metrology and
medical diagnostics. Optical Coherence Tomography (OCT) is a 2-D
extension of OCDR and provides micron-resolution cross-sectional
images of specimens. Most of the industrial and clinical OCDR, OFDR
and OCT machines are disparate, expensive, cumbersome to use,
bulky, not very efficient and are fragile. Everett et al. (2006)
discusses these systems in isolation and in some combinations in
detail.
[0004] However, a more compact and integrated system and an
apparatus would make diagnosis more accurate and the apparatus more
portable.
SUMMARY
[0005] The invention discloses a system, method and apparatus, for
evaluating the retinal microstructure in diabetic patients and
other substructure for failure analysis using optical coherence
domain reflectometry (OCDR), optical frequency domain reflectometry
(OFDR), optical coherence tomography (OCT), Doppler processing and
Doppler OCT technology in combination.
[0006] In one embodiment, an apparatus comprises of a light source,
isolator, beam splitter, optical delivery unit, specimen,
volume-phase holographic grating unit, detector array and a
processor containing novel algorithms for image processing. This is
described as the basic configuration throughout the instant
application with mirror addition and deletion of components. In
another embodiment, the apparatus mentioned above has also at least
one of a faraday rotator mirror, fractional wave mirror, waveplate
(e.g., .lamda./8), a fiber-optic mirror and a free space
mirror.
[0007] In another embodiment as an additional feature, a
polarization compensator is added to the basic configuration
mentioned above. In one embodiment, a fiber stretcher is added in
the basic configuration. The fiber stretcher is used to adjust the
path-length in the corresponding arm of the system.
[0008] In one embodiment, a system comprising of light source,
provides a broad band light for acquiring an image from subsurface
area of a specimen. The specimen may be, but not limited to a
moving sample, a stationary sample or a combination of both. In
another embodiment, the system is modular so that a user can add
off-the-shelf products to enhance the system capabilities. In
another embodiment, several combinations of the basic configuration
and additional components may be added to enhance the performance
of the apparatus as a system as shown in the various figures that
accompany this application, but not limited to only those.
[0009] In another embodiment, a compensating algorithm resides in
the processor to create a superior image. The processor uses the
algorithms such as the frequency resampling, demodulation,
dispersion compensation, and Doppler processing to produce highly
sensitive and high quality images. In another embodiment, the
system performs spectroscopic detection. The resultant spectra are
analyzed by the processor using inverse Fourier transformation and
relevant signal processing for obtaining depth dependent (i.e.
axial) reflectivity profile called A-scan. In another embodiment,
two dimensional tomographic images, B-scan, are created from a
sequence of axial reflectance profiles acquired by scanning the
specimen.
[0010] In one embodiment, a system may comprise of a light source,
isolator, processor, fiber stretcher, source arm, reference arm,
sample arm, detection arm, beam splitter, detector array,
volume-phase holographic grating unit, optical delivery unit,
.lamda./8 waveplate, fractional wave mirror and a specimen for
analysis.
[0011] In another embodiment, the system comprises of a light
source, isolator, processor, detector array, fiber stretcher,
source arm, reference arm, sample arm, detection arm, beam
splitter, volume-phase holographic grating unit, optical delivery
unit, polarization compensator, mirror and a specimen for
analysis.
[0012] In another embodiment, the system comprises of a light
source, isolator, processor, detector array, fiber stretcher,
source arm, reference arm, sample arm, detection arm, beam
splitter, volume-phase holographic grating unit, optical delivery
unit, polarization compensator and another optical delivery system
located on the reference arm, free space mirror and a specimen for
analysis.
[0013] In another embodiment, a system may comprise of a tunable
light source, isolator, processor, fiber stretcher, source arm,
reference arm, sample arm, detection arm, beam splitter, detector,
optical delivery unit, .lamda./8 waveplate, fractional wave mirror
and a specimen for analysis.
[0014] In another embodiment, the system comprises of a light
source, isolator, processor, detector array, fiber stretcher,
source arm, reference arm, sample arm, detection arm, beam
splitter, volume-phase holographic grating unit, optical delivery
unit, polarization compensator located on the sample arm,
fractional wave mirror and a specimen for analysis.
[0015] In one embodiment, a system may comprise of a light source,
isolator, processor, fiber stretcher, source arm, reference arm,
sample arm, detection arm, beam splitter, detector array,
volume-phase holographic grating unit, optical delivery unit,
.lamda./8 waveplate, faraday rotator mirror and a specimen for
analysis.
[0016] In one embodiment, a system may comprise of a tunable light
source, isolator, processor, fiber stretcher, source arm, reference
arm, sample arm, detection arm, beam splitter, detector, optical
delivery unit, .lamda./8 waveplate, faraday rotator mirror and a
specimen for analysis.
[0017] In another embodiment, the system comprises of a light
source, isolator, processor, detector array, fiber stretcher,
source arm, reference arm, sample arm, detection arm, beam
splitter, volume-phase holographic grating unit, optical delivery
unit, polarization compensator located on the sample arm, faraday
rotator mirror and a specimen for analysis.
[0018] In another embodiment, the system enables a user to adjust
the reference arm and the sample arm in order to adjust the
variance of the light beam to get a better quality image.
[0019] In one embodiment, a method of acquiring sensitive, high
quality image of the subsurface for diagnostics and failure
analysis is described.
[0020] In another embodiment, light from a broadband light source
operating at a suitable center wavelength is sent to an isolator,
and then to the beam splitter using the source arm of the
apparatus. In another embodiment, the beam splitter splits the
broadband light into two parts. One part of the light beam goes to
the reference mirror using the fiber stretcher (on the reference
arm) and other beam goes to the specimen using the sample arm.
[0021] In another embodiment, the reflected light from the mirror
using the reference arm joins the light reflected from the specimen
using the sample arm. In another embodiment, the combined light
from the reference arm and sample arm is split again at the beam
splitter and part of the beam goes back to isolator. The other part
of the beam goes to the VPH (volume-phase-holographic grating unit)
and detector array using the detector arm. The beam then is
transformed using novel algorithms that are discussed in detailed
description and transformed to produce high quality, sensitive
images. In another embodiment, if the image quality is poor then
the whole process is repeated, the user is given the option to
adjust the reference arm length, adjust the polarization using
polarization compensation, or focus the light using optical
delivery unit to further improve the image.
[0022] In one embodiment the apparatus comprising of many
components and systems is modular. In another embodiment the
processor/computer-readable media houses the novel algorithm to
process the beam that has the spectrophotometric image information
of the specimen.
[0023] In one embodiment, a method to process the acquired image
using several types of systems using the algorithms residing in the
computer-readable media is described.
[0024] The above mentioned summary presents a simplified version of
one or more embodiments in order to provide a basic understanding
of such embodiments. This summary is not an extensive overview of
all contemplated embodiments, and is intended to neither identify
key or critical elements of all embodiments nor delineate the scope
of any or all embodiments. Its sole purpose is to present some
concepts of one or more embodiments in a simplified form as a
prelude to the more detailed description that is presented later.
Other aspects will be apparent from the following description,
figures and the appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0025] FIG. 1 is a block diagram of an OCDR-OCT system 100, in
accordance with an embodiment of the present invention; the key
novel elements being volume phase holographic grating unit, fiber
optic Fractional Wave mirror, fiber stretcher, and .lamda./8
waveplate.
[0026] FIG. 2 is a block diagram of the OCDR-OCT system 100 similar
to that in FIG. 1 except that the Fractional Wave mirror is
replaced by a fiber optically integrated mirror, and the .lamda./8
waveplate is eliminated and a polarization compensator is
introduced.
[0027] FIG. 3 is a block diagram of the OCDR-OCT system 100 similar
to that in FIG. 2 except that the fiber optically integrated mirror
is replaced by a free space mirror.
[0028] FIG. 4 is a block diagram of the OFDR-OCT system 415 similar
to that in FIG. 1 except that the broad-band source is replaced by
a tunable frequency source, detector array is replaced by a single
high-speed detector, and the diffraction grating is eliminated.
Such a system is called swept-source OFDR/OCT.
[0029] FIG. 5 is a block diagram of the OCDR-OCT system 100 similar
to that in FIG. 1 except the .lamda./8 waveplate is eliminated and
a polarization compensator is introduced in the sample arm.
[0030] FIG. 6 is a block diagram of the OCDR-OCT system 100 similar
to that in FIG. 1 except that the Fractional Wave mirror is
replaced by a fiber optically integrated Faraday Rotator
Mirror.
[0031] FIG. 7 is a block diagram of the OFDR-OCT system 415 similar
to that in FIG. 4 except that the Fractional Wave mirror is
replaced by a fiber optically integrated Faraday Rotator
Mirror.
[0032] FIG. 8 is a block diagram of the OCDR-OCT system 100 similar
to that in FIG. 5 except that the Fractional Wave mirror is
replaced by a fiber optically integrated Faraday Rotator
Mirror.
[0033] FIG. 9 is a flow chart of describes a method of acquiring an
image from a specimen using the OCDR-OCT system.
[0034] FIG. 10 is a flow chart of method of using the
apparatus.
[0035] FIG. 11 is a flow chart of overview of methods of the
signals and images being processed from the start to finish.
[0036] FIG. 12 is a flow chart of method of demodulating the signal
to recover the complex envelope of the OCT/OCDR/OFDR signal.
[0037] FIG. 13 is a flow chart of method of Doppler processing the
signal to estimate the Doppler shift and the corresponding
velocities of the particles in the specimen.
[0038] Other features of the present embodiments will be apparent
from the accompanying figures and from the detailed description
that follows.
DETAILED DESCRIPTION
[0039] The instant disclosure describes a technological advancement
of acquiring an image that is stationary, moving and/or combination
of stationary and moving specimen in subsurface area and enhancing
the quality of the image by using proprietary algorithms. The
disclosure also describes an apparatus, a system and a method for
evaluating the retinal microstructure in diabetic patients and
other substructure for failure analysis using optical coherence
domain reflectometry (OCDR), optical frequency domain reflectometry
(OFDR), optical coherence tomography (OCT), Doppler processing and
Doppler OCT technology in combination.
[0040] OCDR-OCT System: FIG. 1 shows an OCDR-OCT system 100
comprising of a light source 105, isolator 121, processor 114,
fiber stretcher 112, source arm 101, reference arm 102, sample arm
103, detection arm 104, beam splitter 106, detector array 110,
volume-phase holographic grating unit 113, optical delivery unit
108, .lamda./8 plate 111, Fractional wave mirror 109 and a specimen
107 for analysis. This is one of the preferred embodiments for our
invention.
[0041] A light source 105, in a system or as a part of the
apparatus, may comprise of off-the-shelf light sources.
[0042] The center wavelength (.lamda..sub.0) most ideal for the
retinal applications range from 750 nm till 1050 nm. Water (and
aqueous humor) absorption is minimal for this wavelength range. The
power for retinal applications ranges from 0.1 mW to 10 mW. Per
ANSI safety standards only 0.75 mW are permitted incident on the
eye at this wavelength range of 750 nm till 1050 nm. The center
wavelength most ideal for the non-retinal applications (e.g., skin,
anterior segment of the eye, gastrointestinal tract, lungs, teeth,
blood vessels, subsurface area of semi-conductors, chip
manufacturing, sensitive medical equipment's etc.) range from 1050
nm till 1350 nm. The longer wavelength is more suitable for thick
scattering tissues since scattering is less at higher wavelengths.
The system depth resolution (DR) is inversely proportional to the
FWHM spectral width (or bandwidth.DELTA..lamda.). It is given by
the following equation:
DR = 2 ln 2 .pi. .lamda. 0 2 .DELTA..lamda. ( Eq 1 )
##EQU00001##
[0043] The full-width-half-max (FWHM) spectral width of the light
source typically ranges from 10 nm till 150 nm. The power for
non-retinal applications ranges from 0.1 mW till 30 mW in the
wavelength range from 1050 nm till 1350 nm. The full-width-half-max
(FWHM) spectral width of the light source typically ranges from 10
nm till 150 nm.
[0044] The light source 105 may be electrically operated. These can
be battery operated while in transit. The forward voltage typically
ranges from 2 to 10 Volts. The forward current typically ranges
from 100 mA to 1 A. Some of these sources need to be
thermo-electrically controlled (TEC). The operating internal
temperature for some sources is typically 25.degree. C. The
corresponding thermistor resistance is 10 kilo-Ohms (10 k.OMEGA.).
Typical TEC current is 1.5 A. Typical TEC voltage is 3-4V. The
light source may also be tunable light source as shown in other
system/apparatus embodiments.
[0045] The isolator 121 protects the light source from back
reflections and permits the transmission of light in the forward
direction with a limited loss. The fiber-optic isolator used in
idevice would need to operate on a broad range of spectrum to cover
the full spectral-width of the light source (Depending upon the
source spectral shape, typically 2* FWHM bandwidth .DELTA..lamda.).
Thus the operating wavelength range is
.lamda..sub.0+/-.DELTA..lamda.. Typical isolation is 20-40 dB, and
insertion loss is 0.5-3 dB. The polarization dependent loss is
typically 0.5 dB or less. Return loss is typically more than 40
dB.
[0046] The isolator 121 comprises of an input linear polarizer, a
(.lamda./8) Faraday rotator or a waveplate, and an output linear
polarizer. The (.lamda./8) Faraday rotator or a waveplate rotates
the light transmitted through the input polarizer by 45 degrees.
The output polarizer needs to have the same direction as "the input
polarizing direction rotated by 45 degrees" in order to have the
maximum transmission and maximum isolation. The light returning to
the isolator from the remaining system gets linearly polarized by
the output polarizer and is rotated by 45 degrees, making it
orthogonally polarized as compared to the input polarizing
direction. Thus, the returning light is totally absorbed.
[0047] Fiber stretcher 112 consists of a fiber looped around a
piezoelectric device (which is a solid block that can be expanded
or contracted by electric voltage). The purpose of a fiber
stretcher is to increase or decrease the path-length in the
interferometer that is on the detection arm by increasing or
decreasing the fiber-length. Although the fiber stretcher 112 is
shown in the reference arm, it can be placed ether in the reference
arm or sample arm. If the fiber stretcher 112 is kept in the
reference arm, since the fiber is looped around the piezoelectric
device, care must be taken to provide extra fiber in the sample arm
so that the sample arm and reference arm path lengths are
matched.
[0048] The fractional wave mirror 109 consists of a fiber-optic
mirror preceded by a fractional [45 degrees (.lamda./8)] waveplate.
The polarization of light incident on the wave plate is rotated by
45 degrees, and is directed to the mirror. The reflected light is
further rotated by 45 degrees by the fractional [45 degrees
(.lamda./8)] waveplate and hence the resulting polarization is
orthogonal to the incident polarization. We would use a fiber
optically integrated birefringent reference mirror is at least one
of fractional wave mirror, mirror, free space mirror and Faraday
rotator mirror. A modified formula based on LeFvre is disclosed in
this disclosure and which is as follows:
[0049] Mechanical stress on the fiber is causes birefringence in
the fiber. Stress can be generated by simply bending the fiber.
According to LeFevre (U.S. Pat. No. 4,615,582), the fractional wave
plate can be built by looping the fiber into N loops having a
radius R. The refractive index difference .DELTA.n for two
orthogonal polarizations is given by
.DELTA. n = b ( r R ) 2 ( Eq 2 ) ##EQU00002## [0050] b is a
constant depending upon the photoelastic coefficient of the fiber,
r is the radius of the fiber and R is the radius of the fiber loop.
Thus, if we want to create a .lamda./m (where m is an integer)
waveplate, which will introduce a path-length shift of .lamda./m
between 2 polarizations, we'll need to create a loop of fiber
length L to create the path-length shift of .DELTA.nL. However,
since the length of the fiber is also equal to 2.pi.NR, where N is
the number of loops, we get
[0050] ( 2 .pi. NR ) b ( r R ) 2 = .lamda. m or ( Eq 2 ) R = ( 2
.pi. mN ) b r 2 .lamda. ( Eq 4 ) ##EQU00003## [0051] To create a
fractional wave plate of
[0051] .lamda. 8 , ##EQU00004## [0052] and N=1 (single loop),
b=0.25, m=8, r=125 microns, .lamda.=0.8 microns, we get
[0052] R = ( 2 .pi.8 ) 0.25 ( 125 ) 2 0.8 = 5 .pi. * 15625 = 24.54
cm ( Eq 5 ) ##EQU00005##
Please note that a (2M+1).lamda./m waveplate where M is an integer
between -.infin. to .infin. will have a similar effect as a
.lamda./m waveplate.
[0053] In typical state-of-the-art OCT systems, light exits a fiber
tip in the reference arm and the light returns from a retro
reflecting mirror mounted in air. This increases system complexity
and bulkiness. In some embodiments of instantly described
invention, a fiber-optically integrated fractional wave mirror 109
in the reference arm 102 of the OCDR-OCT system 100 can be used.
Since the polarization of the retro reflected light is orthogonal
to the incident light, fiber birefringence effects effectively get
cancelled in the reference arm 102.
[0054] Detector array 110 is a line-scan camera. It has typically
1024-4096 pixels, though the proposed embodiment is not limited to
these numbers. Typically it is a CCD or CMOS camera. Line-rate
(rate of acquisition of arrays) is typically 10000 lines/s to
400000 lines/s, though the proposed embodiment is not limited to
these numbers. Each pixel outputs a value which typically has an
8-bit or 12-bit format, though the proposed embodiment is not
limited to these numbers. The pixel size is typically 14 microns
(height) and 14 microns (width). The light dispersed by the grating
is focused on the detector array. The output of the array
(line-scan camera) is typically directed to the computer using an
Ethernet cable (e.g., Gigabit Ethernet) or a USB (typically 2.0 or
3.0) cable, etc. The operating wavelength ranges from 400 nm to
1100 nm for retinal applications. The above numbers and examples
are given for illustrative purposes only, the proposed embodiment
is not limited to these numbers or examples.
[0055] The beam splitter 106 (made of fiber optics) splits the
light typically into 50/50. It is built using two fused single-mode
fibers. The fiber for retinal applications (.about.800 nm
wavelength) has 4-6 microns core diameter and 125 microns cladding
diameter, 0.130 core numerical aperture (NA), cutoff wavelength of
typically 730 nm. The insertion loss (in addition to designed 3 dB
or 50% loss) is typically 0.3 dB. For the couplers used for OCT,
the length of the fiber in the reference and sample arms is very
important and the lengths are specified with tight tolerances.
[0056] The waves reflected back from the sample arm 103 and the
reference arm 102 interferes at the detector array 110. Since the
interference signal is only created when the polarization in the
reference arm 102 matches with that in the sample arm 103, in some
embodiments, one can include by way of example but not by
limitation a 45 degrees .lamda./8 waveplate 111 in the sample arm
103 just before the light is incident on the optical delivery unit
108. Since the polarization of the retro reflected light will be
almost orthogonal to the incident light (considering the fact that
the birefringence in the specimen 107 will modify the polarization
state), the birefringence effects in the sample arm fiber 103 of
the interferometer 100 will get cancelled. In a preferred
embodiment, the .lamda./8 waveplate 111 is constructed using fiber
optic components.
[0057] In an embodiment of this invention, other waveplates (non-45
degrees Faraday rotators) can be used. The quality of the
interferometric signal (e.g., contrast and signal to noise ratio)
will be better or worse depending upon the polarization properties
of the specimen in the sample arm.
[0058] The instant system and apparatus that comprises of Optical
coherence tomography (OCT) and OCDR that is very similar to
ultrasound imaging. OCDR-OCT provides cross-sectional images of
micro-features that are acquired from adjacent depth resolved
reflectivity profiles of the tissue. OCT also employs a fiber
optically integrated Michelson interferometer illuminated with a
short coherence length light source such as a superluminiscent
diode (SLD). The interferometric data are processed in a
processor/computer and displayed as a gray scale image. In an
OCDR-OCT image, the detectable intensities of the light reflected
from human tissues range from 10.sup.-5 to 10.sup.-11th part of the
incident power.
[0059] OCDR-OCT system 100 and OFDR-OCT 415 are able to image
sub-surface retinal microstructure and has been useful for
diagnosis and management of diabetic retinopathy. Abnormalities in
blood-flow circulation due to diabetes are the root cause behind
retinal microstructure damage. However, no clinical tools exist
that can perform functional and velocity mapping of blood vessels
in the retina for tracking early development of diabetic eye
diseases. Therefore, there is a need for an automated, low-cost and
compact tool based on Doppler OCT for tracking progression and
management of diabetic retinal diseases by performing 3-D
functional mapping of blood circulation in the retina. Such a
device will be extremely useful in detecting earliest signs of
diabetic retinopathy and hence it will be an ideal tool for
screening diabetic patients at risk of developing retinopathy.
Since it has been proven that glucose and blood-pressure control
are the best methods for managing diabetic retinopathy, instant
Doppler OCT system will be an ideal low-cost tool, which will
permit screening as well as management for the disease. The
invention presented here provides such a system and addresses these
issues.
[0060] In another preferred embodiment, the .lamda./8 waveplate 111
is a fractional-waveplate constructed using fiber optic components.
It would be constructed in the optical delivery unit near the end
of the fiber segment in the optical delivery unit. Fractional
waveplate 111 is located on the sample arm of the apparatus. It may
be made an integral part of the optical delivery 108. The
fractional wave mirror 109 consists of a fiber-optic mirror
preceded by a fractional [45 degrees (.lamda./8)] waveplate. The
polarization of light incident on the waveplate is rotated by 45
degrees, and is directed to the mirror. The reflected light is
further rotated by 45 degrees by the fractional [45 degrees
(.lamda./8)] waveplate and hence the resulting polarization is
orthogonal to the incident polarization. In another embodiment, a
free-space-bulk 45 degrees (.lamda./8) wave plate is used at the
end of the optical delivery unit.
[0061] Instant OCDR-OCT system uses spectroscopic detection method.
Basically the interferometric light exiting the detector arm 103 is
dispersed via a grating. The spectra are acquired using a line-scan
camera. The resulting spectra are typically (by way of example, not
by limitation) transferred to a processor for inverse Fourier
transforming and relevant signal processing (such as obtaining the
complex envelope of the interferometric signal) for obtaining depth
dependent (i.e., axial) reflectivity profiles (A-scans). The axial
resolution is governed by the source coherence length, typically
.about.3-10 .mu.m. Two dimensional tomographic images (B-scans) are
created from a sequence of axial reflectance profiles acquired
while scanning the probe beam laterally across the specimen or
biological tissue.
[0062] A-scan: A-scan is a plot of reflectivity of scatterers and
layers as a function of depth at a given lateral location. It is
computed as follows:
a) The interferometric light exiting the detector arm is dispersed
via a grating. b) The dispersed light is a spectrum which is
focused on a detector array or a line-scan camera. c) The recorded
spectra are typically transferred to a processor d) An inverse
Fourier transform of the spectrum is computed e) Relevant signal
processing is performed (such as removing the duplicate data and
strong spikes at the center of the inverse Fourier transform) f)
The resulting arrays is a depth dependent (i.e., axial)
reflectivity profiles (A-scans). g) The axial resolution is
governed by the source coherence length, typically -3-10 .mu.m.
[0063] B-scan: Two dimensional tomographic images (B-scans) are
created from a sequence of axial reflectance profiles acquired
while scanning the probe beam laterally across the specimen or
biological tissue. The following are detail steps: [0064] a) An
A-scan is acquired at a given lateral location. [0065] b) A mirror
is scanned using a scanner such as a galvanometer in the optical
delivery unit [0066] c) Multiple A-scans are acquired at various
lateral locations. [0067] d) A matrix is generated where columns
indicate different lateral locations and rows indicate reflectivity
at each depth in each A-scan [0068] e) The matrix is displayed as
an image, which is also a B-scan
[0069] Processor comprises of many algorithms that are discussed
below. There may be a combination of algorithms that may be used
for image formation. The algorithms may be used individually or in
certain sets, or in a serial manner. FIGS. 11, 12 and 13 discuss
some embodiments, but the use is not limited to that only.
[0070] Dispersion compensation is an algorithm used in the instant
invention. Dispersion is caused by mismatch in the materials in the
reference and sample arms. In many situations, light may travel
through more fiber in the reference arm and more air in the sample
arm. This is especially possible if we use a fiber-optic mirror,
which would result in no air in the reference arm. There would be
some air in the sample arm as light needs to travel through optical
delivery unit and focus on the specimen. The result of dispersion
is loss of resolution and distortion of signal in the A-scan. Hence
it needs to be corrected using dispersion algorithm. The corrected
signal will have better depth resolution and higher fidelity.
[0071] In some embodiments, another way of achieving the
polarization matching is to use a polarization compensator 220 as
shown in FIG. 5 instead of using .lamda./8 waveplate 111. In other
embodiments, combinations of waveplates and polarization
compensators can be used to achieve the desired polarization
matching. The prior art the control for the fiber optic
polarization interferometer does not specify the precise location
of the fiber-optic wave-plates along the reference or sample
arm.
[0072] In the prior art, OCT systems need to dynamically adjust
polarization (before each patient exam) in the sample arm 103 in
order to match with polarization in the reference arm. We will not
need dynamic polarization compensation due to instantly described
novel approach.
TABLE-US-00001 TABLE 1 Advantages of Faraday rotator mirror Sr.
Faraday Rotator mirror advantage Implications No. compared to
mirror mounted in air for OCT-OCDR [1] Polarization effects get
cancelled due to the Polarization orthogonal polarization of the
retroreflected insensitivity, no light need for dynamic
compensation [2] Easy to assemble, no alignment needed in the Low
cost of reference arm production [3] Integral Part of the 3-dB
coupler and reference Robust, rugged, arm assembly (wherein the
same fiber is used compact, low-cost to build the fiber optic
splitter and the Faraday rotator mirror in the reference arm
fiber.)
[0073] Volume-Phase Holographic (VPH) Gratings: In the prior art,
clinical OCT systems use ruled gratings for dispersing light on a
line-scan camera in the detector arm. Ruled gratings are cumbersome
and expensive. In some embodiments of currently described
embodiment, volume-phase holographic (VPH) grating unit 113, which
is essentially a transmission grating with alternating refractive
indices, can be used. VPH grating unit are highly efficient,
compact, rugged, and low-cost at telecom wavelengths since these
are widely used in telecom industry. VPH grating unit were first
developed for astronomy applications. The benefits of VPH grating
unit are explained as follows (Table 2):
TABLE-US-00002 TABLE 2 Advantages of VPH grating unit: Implications
for OCT Sr. No. VPH grating advantage compared to ruled grating and
OCDR [1] have very high diffraction efficiency approaching 100%.
high sensitivity [2] Polarization effects are not as bad as in
ruled gratings, high sensitivity [3] lack many anomalies apparent
in ruled gratings. High image quality [4] Ghosting and scattered
light from a VPH grating is high substantially reduced compared to
ruled gratings. sensitivity [5] Can be tuned to shift the
diffraction efficiency peak to a high desired wavelength.
sensitivity [6] Can be tuned to direct more energy into higher
diffraction high orders using non-sinusoidal refractive index
modulation on sensitivity the grating (Barden et al. 2000); a
versatility not possible with classical gratings. [7] have high
line densities (<6000 lines/mm) than ruled gratings Higher scan
at a lower cost depth, lower cost [8] can be cleaned due to the
encapsulated nature of the grating. More life, lower cost, higher
sensitivity [9] The encapsulated nature permits antireflection
coatings on lower cost, the surfaces of the grating. higher
sensitivity [10] can be designed to work in the Littrow
configuration (as Lower cost described in (Barden et al. 2000),
where the fringe structure to is normal to the grating surface, and
the grating will have no manufacture anamorphic magnification at
the Bragg wavelength), resulting in a simplification of the
line-scan camera objective optics (auto-collimated entrance and
exit beams and the same focal length objectives can be used).
[0074] In some embodiments of this invention, the grating disperses
light and a lens focuses it into a detector array 110. By way of
example, but not by limitation, this array can be a line-scan
camera, which has quantum efficiency p at the operating
wavelengths. The resulting data set is inverse Fourier transformed,
processed in a processor 114 and displayed as a gray scale or
pseudo-color image. By way of example, not by limitation, this
processor can be a computer, off-the-shelf integrated circuit,
Field application specific integrated circuit (ASIC), Programmable
Gate Array (FPGA), a graphical processing unit (GPU) an embedded
system or a microcontroller.
TABLE-US-00003 TABLE 3 Advantages of fiber optic waveplate coupled
at the end of the fiber in the optical delivery unit Fiber optic
waveplate Sr. coupled at the end of Implications for OCDR- No. the
fiber in the optical delivery unit OCT [4] Polarization effects get
Polarization insensitivity, no cancelled due to the need for
dynamic orthogonal polarization of the light compensation
backscattered from the sample [5] Easy to assemble, no alignment
Low cost of production needed in the sample arm [6] Part of the
3-dB coupler and Robust, rugged, compact, sample arm assembly
low-cost
TABLE-US-00004 TABLE 4 Advantages of fiber optic waveplate coupled
before the optical delivery unit verses placing the waveplate
before the sample Fiber optic waveplate Sr. coupled before the
Placing the waveplate No. optical delivery unit before the sample
[7] Robust, rugged, compact, low-cost Fragile, bulk, expensive [8]
Easy to assemble, no alignment Hard to assemble, alignment needed
in the sample arm needed in sample arm [9] Part of the 3-dB coupler
and Part of the optical delivery sample arm assembly unit
[0075]
.lamda.)k=1/.lamda.).delta..lamda.=(.lamda.max-.lamda.min)/Nkmax=1/-
.lamda.minkmin=1/kmax.delta.k=(kmax-kmin)/N.delta.k=(kmax-kmin)/XNS.sub.cc-
d(k.sub.0)=S.sub.ccd(k.sub.l)+U.sub.0[S.sub.ccd(k.sub.u)-S.sub.ccd(k.sub.l-
)];
U 0 = k 0 - k l k u - k l S ccd ( k l ) = W ccd ( .lamda. l , x ) S
ccd ( k u ) = W ccd ( .lamda. u , x ) ##EQU00006##
Alternate embodiments of instantly described OCT-OCDR system
invention:
[0076] FIG. 2 is a block diagram of a system similar to that in
FIG. 1 except that the Fractional Wave mirror is replaced by a
fiber optically integrated mirror 217, and the (.lamda./8)th
waveplate is eliminated and a polarization compensator 220 is
introduced. FIG. 2 has standard fiber-optic-mirror in the reference
arm, which still permits use of novel algorithms such as frequency
resampling, demodulation, dispersion compensation, and Doppler
processing algorithms. Polarization compensator 220 is also known
as Fiber optic polarization compensators. These are based on the
principles of fractional wave plate above. It consists of 3 coils
of fiber on 3 different paddles arranged in a series. The first
fiber coil is a quarter wave plate, the second fiber coil is a half
wave plate (typically the fiber is looped around twice for the same
paddle diameter as the first paddle), the last fiber coil is a
quarter wave plate. These 3 paddles can be rotated freely with
respect to each other to produce any polarization state.
[0077] There is another type of polarization compensator, which
applies pressure to the fiber to create birefringence. The slow
axis is in the direction of the pressure applied. This fiber
squeezer can be rotated around the fiber to rotate the direction of
the slow axis. Thus, any arbitrary polarization can be created.
[0078] Different types of gratings: Volume-Phase Holographic 113
grating unit is a transmission grating and the diffraction is
achieved by periodic modulation of the refractive index. A similar
effect could be achieved by periodic modulation of grating
substrate thickness instead of (or in addition to) refractive index
modulation.
[0079] Extensions of the proposed interferometer: An
interferometric 2D imaging system (Optical coherence tomography or
OCT) can be constructed using the proposed interferometric system
where the 2D images are obtained by laterally scanning the beam
incident on the sample using a 1-D scanning mirror (which is a part
of the optical delivery unit). An interferometric 3D imaging system
can be constructed using the proposed interferometric system where
the 3D data-sets are obtained by 2D laterally scanning the beam
incident on the sample using a 2-D scanning mirror (which is a part
of the optical delivery unit).
[0080] Both the 2D imaging systems and 3D imaging systems can be
adapted for ophthalmic imaging by using a lens assembly (which is a
part of the optical delivery unit) to focus the light on the
retina.
[0081] Both the 2D imaging systems and 3D imaging systems can be
adapted for an endoscopic or catheter imaging system where the
light in the sample arm is delivered through an endoscope. Thus,
the sample arm fiber passes through an endoscope or a catheter. An
example endoscopic OCT is shown in reference 4 (not as a
limitation), but other endoscopic/catheter systems could be
used.
TABLE-US-00005 TABLE 5 Advantages of instantly described proposed
OCDR-OCT system: Proposed feature in State-of-the-art Sr.
instantretinal OCT Advantage to clinician and clinical retinal No.
machine patient OCT machines [1] Scalable, price goes down
Increased affordability with Price does not go with increasing
sales device adaptation down with volume due to use of device
increasing sales and packaging technologies volume due to use of
labor intensive bulk technologies. [2] Portable Can be easily
transported to Not portable remote localities [3] Rugged and Robust
Can operate in rural Fragile, not robust challenging environment
[4] Use of volume holographic Lower cost, compact, rugged Ruled
grating phase grating [5] Faraday rotator mirror in Lower cost,
compact, rugged Glass mirror reference arm mounted in air [6]
Dynamic polarization Ease of use, patients and Dynamic control not
needed due to clinicians save valuable time polarization control
Faraday mirror above. needed.
[0082] An example lens assembly is described below (not as a
limitation), but other lens assemblies could be used. The OCDR-OCT
system can be adapted to measure retina by collimating the beam
exiting the sample arm fiber, expanding the beam using a lens,
shrinking the beam to project on the cornea, and the cornea and
lens system of the eye will automatically focus the beam on the
retina.
[0083] In another variation of this embodiment (FIG. 3), the fiber
optically mirror can be replaced by a free space mirror 318. The
light can be delivered to the mirror using optical delivery unit
319. FIG. 3 has standard free-space-mirror 318 in the reference
arm, which still permits use of instant algorithms such as
frequency resampling, dispersion compensation, and Doppler
processing algorithms.
[0084] Frequency Domain OCT or Optical Frequency Domain
Reflectometry: In some OCT systems such as frequency domain OCT or
Optical Frequency Domain Reflectrometry (OFDR), the broad-band
light source is replaced by a tunable frequency light source. The
detector array is replaced by a single detector. The use of VPH is
not needed for this invention. In this embodiment of instant
embodiment (FIG. 4), a fiber-optically integrated Fractional wave
mirror 109 in the reference arm 102 of the OFDR-OCT system 415 can
be used. Since the polarization of the retro reflected light is
orthogonal to the incident light, fiber birefringence effects
effectively get cancelled in the reference arm 102. FIG. 4 is
another preferred embodiment of instant invention, which applies
the use of fractional wave mirror 109, and fiber optic .lamda./8
waveplate 111 for optical Frequency domain reflectometry (OFDR)-OCT
system 415 and method. Tunable light source 402 in this embodiment
is applicable to FIGS. 4 and 7 only. The center wavelength most
ideal for the retinal applications range from 750 nm till 1050 nm.
The wavelength of the source is tuned very rapidly (e.g., at a rate
of 10 kHz 1 MHz) within a spectral range of typically 10 to 100 nm
around the center wavelength. The average power of such a source
typically ranges from 0.1 mW to 20 mW depending upon the
applications. The source may be electrically operated. The existing
commercially available sources operate on 110/220V 50/60 Hz power
input. In future, these could be operated using lower voltages and
battery operated while in transit. ADC 424 is added so that the
electrical current is transformed.
[0085] In this embodiment there is no VPH 113 and detector array.
Instead a Detector 422 is added. It is a photo-diode (which
converts light into electricity). The detectors for 300-1000 nm are
typically made up of silicon. The detectors for 900-1700 nm are
typically made up of InGaAs. These are high-speed detectors with
typically 0 to a few hundred MHz bandwidth. It is typically
followed by a high-speed A/D (analog to digital) converter, e.g.,
8-bit or 12-bit with a conversion rate of 1 to 500 Mega
Samples/second. Typical responsivity of photodiodes is 0.1-1 mA/mW.
The output voltages are typically -5 to 5V, with typical 50.OMEGA.
impedance. These assist in achieving typical line-rates (rate of
acquisition of A-scans) of 10000 lines/s to 40000 lines/s. The
output of the A/D converted is typically directed to the computer
using an Ethernet cable (e.g., Gigabit Ethernet) or a USB
(typically 2.0 or 3.0) cable, or directly attached to a computer's
PCI (Peripheral Controller Interface) bus etc.
[0086] FIG. 5 is another embodiment where everything is the same as
FIG. 1, except the .lamda./8 waveplate is replaced by the
polarization compensator (PC) 220. This is a simplified system.
[0087] FIG. 6 is an embodiment similar to that in FIG. 1 except
that the Fractional Wave mirror 109 is replaced by a fiber
optically integrated Faraday Rotator Mirror 609, which is an
off-the-shelf part. Faraday rotator mirror: The device consists of
a fiber-optic mirror preceded by a 45 degrees (.lamda./8) Faraday
rotator. The polarization of light incident on the Faraday rotator
is rotated by 45 degrees, and is directed to the mirror. The
reflected light is further rotated by 45 degrees by the Faraday
rotator and hence the resulting polarization is orthogonal to
incident polarization. We will use the term fiber optically
integrated birefringent reference mirror to indicate a fiber
optically integrated Faraday Rotator Mirror or a fractional wave
mirror.
[0088] A Faraday rotator mirror 609 consists of a magnet. It
changes the polarization of light by Faraday effect. The
polarization of light is affected in the presence of a magnetic
field if it is applied parallel to the direction of propagation.
Therefore, a Faraday rotator consists of a magnet to generate
magneto-optical effect. A Faraday rotator works because one of the
components of polarization of propagating light is in ferromagnetic
resonance with the material, which causes the phase velocity of the
resonating polarization to be higher than the phase velocity of the
corresponding orthogonal polarization.
[0089] FIG. 7 is an embodiment similar to that in FIG. 4 except
that the Fractional Wave mirror 109 is replaced by a fiber
optically integrated Faraday Rotator Mirror 609, which is an
off-the-shelf part. A detector 422 is added.
[0090] FIG. 8 is an embodiment similar to that in in FIG. 5 except
that the Fractional Wave mirror 109 is replaced by a fiber
optically integrated Faraday Rotator Mirror, which is an
off-the-shelf part.
Method of Image Acquisition and Analysis
[0091] FIG. 9 describes a method of acquiring an image from a
specimen using the OCDR-OCT system. A light source may be a tunable
light source, a broadband source, a laser. An apparatus or system
is used to send a specific bandwidth light from a light source to a
specimen 904 using a source arm and sample arm. A backscattered
light from the specimen is received 906 by the optical delivery
unit and/or .lamda./8 plate in one embodiment. An image is formed
908 after going through the VPH and detector array and checked for
quality 910. If the image quality is poor 912 the steps from 904
are repeated. If the image quality is good 914 data is further sent
to produce an image for analysis 916 using the processor
algorithms. The process ends once the image is formed 918.
[0092] FIG. 10 describes the steps of light travelling through the
source to the specimen and the signal from the light being
processed. Light is being delivered using a light source using the
sample arm to the beam splitter 1004. Beam splitter splits the
light into two parts sending the first path light to reference arm
1008 and second path light into the sample arm 1010. The second
path light goes to the specimen via the optical delivery unit. The
specimen in this case may be retina of the eye for a diabetic
patient. Since the blood flows at irregular intervals and the
picture is not static at times; stationary-object
light-backscattering, moving-object-light-backscattering and
combined-object-light-backscattering is returned to the beam
splitter.
[0093] Sample arm sends the second path of light to the specimen
and the specimen reflects back the second path of light as a
returning light via the optical delivery unit and the waveplate to
the beam splitter 1014. A fiber optically integrated birefringent
mirror (reference mirror) returns the light into the fiber to be
combined with the returning light from the specimen at the beam
splitter 1016. The combined light splits in the beam splitter again
to go into source and detector arms 1018. A partial returning light
from the beam splitter travels through a detector arm to a
volume-phase holographic grating unit and a detector array in
OCDR-OCT system or enters the detector if it is OFDR-OCT system to
be converted to digitized signal 1020. Digitized signal enters the
processor for image formation 1022. The method ends there 1024. On
the other hand partial light returns to the isolator using the
source arm 1026 and the method ends there 1028.
[0094] FIG. 11 shows a high level flow of the processing
algorithms. Step 1102 is the beginning step. For the OCDR-OCT
system, the spectra are acquired from the detector array as
explained earlier (Step 1104). Since the acquired spectra are
typically spaced in equal intervals of wavelength, in the step
1106, the spectra are resampled at equal intervals of spatial
frequency (k-space) using a frequency resampling algorithm. Next in
step 1108, demodulation, which includes inverse Fourier
transforming, is performed to extract the complex envelope of the
signal. Next in order to correct for the dispersion in the system,
the dispersion compensation is performed in step 1110. Next in step
1112, Doppler processing is performed to extract velocity images.
The method ends in step 1114. These algorithms are processed in a
processor 114 and displayed as a gray scale or pseudo-color image.
By way of example, not by limitation, this processor can be a
computer, Field Programmable Gate Array (FPGA), an embedded system
or a microcontroller.
[0095] Frequency Resampling:
[0096] The spectra W.sub.ccd(.lamda.,x) measured by the
spectrometer (i.e., the output of the digital array) are equally
spaced in wavelength (.lamda.). However in order to obtain an
accurate A-scan measurement by inverse Fourier transforming, the
spectra need to be re-measured at equal intervals of spatial
frequency (k=1/.lamda.). Thus, if N is the total number of samples,
the spectra are measured at equal intervals in wavelength
.delta..lamda.=(.lamda.max-.lamda.min)/N. The spectra need to be
equally spaced in k-space. Thus, if the corresponding maximum and
minimum wavenumbers are kmax=1/.lamda.min and kmin=1/kmax, then the
spectra need to be re-sampled at equal intervals in k given by
.delta.k=(kmax-kmin)/N to obtain S.sub.ccd(k,x). If the data are
over-sampled while re-sampling by a factor of X, then
.delta.k=(kmax-kmin)/XN.
[0097] There are many algorithms for re-sampling the spectra. One
such method is simple linear interpolation as described by
[Vergnole et al 2010]. Thus, if we need to calculate the spectrum
S.sub.ccd(k.sub.0,x) at a location k.sub.0, and the spectra are
measured at the nearest neighboring wavenumbers k.sub.u (upper
wavenumber=1/.lamda..sub.u, .lamda..sub.u is the upper wavelength),
k.sub.1 (lower wavenumber=1/.lamda..sub.1, .lamda..sub.1 is the
lower wavelength) Then
S.sub.ccd(k.sub.0)=S.sub.ccd(k.sub.l)+U.sub.0[S.sub.ccd(k.sub.u)-S.sub.cc-
d(k.sub.l)];
U 0 = k 0 - k l k u - k l ##EQU00007##
and note that S.sub.ccd(k.sub.l)=W.sub.ccd(.lamda..sub.l,x)=and
S.sub.ccd(k.sub.u)=W.sub.ccd(.lamda..sub.u,x)
[0098] Another method described by [Vergnole et al. 2010] is spline
interpolation. A preferred and faster method of interpolation is
achieved by convolution using a Kaiser-Bessel window as described
by [Vergnole et al. 2010].
S.sub.ccd(k.sub.0)=.SIGMA..sub.l=-M/2.sup.M/2 S.sub.ccd(k.sub.l)
C.sub.0(k.sub.l) where k.sub.1 are the non-linearly placed
neighboring values of wavenumbers, M is the size of the convolution
kernel. M can be any value, however a value between 3 to 9 can
yield good results.
C 0 ( k l ) = I 0 ( .gamma. 1 - ( 2 H M ) 2 ) M ##EQU00008## where
##EQU00008.2## H = smaller of M 2 or ( k - k l ) / .delta. k
##EQU00008.3##
and I.sub.0 is the zero-order Bessel function of the first kind. To
the best of our knowledge, this is the first time a convolution
based interpolation method is used for the OCDR/OFDR/OCT system in
which, the polarization issues are solved by using a fiber
optically integrated birefringent mirror in the reference arm.
[0099] Next in FIG. 12, we present novel algorithm such as
demodulation algorithm (step 1202), which is also insant version of
the modified Hilbert transform algorithm: [0100] 1) Resampled CCD
spectra S.sub.ccd(k,x) are obtained as a function of k (wavenumber)
and lateral dimension x (step 1204). [0101] 2) Spectra are Fourier
transformed in lateral dimension to obtain spectra P.sub.ccd (k,u)
where u is frequency in lateral dimension (step 1206). [0102] 3)
The negative frequency signals are zeroed out using Heaviside
function H(u) to provide P'.sub.ccd (k,u) (step 1208). [0103] 4)
The P'.sub.ccd (k,u) is inverse Fourier transformed to obtain
complex spectra S'.sub.ccd (k,x) (step 1210). [0104] 5) S'.sub.ccd
(k,x) is inverse Fourier transformed in k (i.e., depth) dimension
to obtain complex envelop in Eq. 2 (step 1212)
[0104] s(z,x)=A(z,x)exp[-j(2.pi.f.sub.s(z,x)zT/D+.phi.(z,x))]. (Eq
6)
[0105] Here A(z,x) is the amplitude of the detected signal
corresponding to the depth-resolved reflectivity obtained in
conventional OCT imaging and .phi.(z,x) is the phase corresponding
coherent interference of backscattered waves, commonly known as
speckle. Here z is the depth location, x is the lateral location, D
is total depth of A-scan, T is the time taken to acquire an A-scan.
For a broadband source, A(z,x) is a highly localized function
(e.g., a Gaussian) whose width determines the axial resolution of
the OCT image. f.sub.s is Doppler shift in light backscattered from
moving objects in the sample. A scatterer in the sample moving with
a velocity V.sub.s induces a Doppler shift in the sample arm light
by the frequency
f.sub.s=2V.sub.s[cos .theta.]n.sub.tv.sub.0/c (Eq. 7)
where .theta. is the angle between the sample probe beam and the
direction of motion of the scatterer, n.sub.t is the local tissue
refractive index, v.sub.0 is the source center frequency, and c is
the light velocity.
[0106] Dispersion compensation: Group velocity dispersion needs to
be matched between the reference and sample arms irrespective of
using the Faraday rotating mirror. In some embodiments of instant
invention, dispersion is compensated numerically by flattening the
Fourier domain phase of a mirror reflection as explained in
[Kulkarni 1999]. Current proposed procedure comprises of: [0107] a)
Measuring the interferogram by placing a mirror in the sample,
computing the complex envelope
m.sub.s(z)=A.sub.m(z)Exp(j.phi..sub.m(z)) [Here z is distance in
depth, A.sub.m is amplitude and .phi..sub.m is phase) for the
interferogram as described in Kulkarni (1999). [0108] b) Computing
the complex envelope for each interferogram measurement for any
desired specimen as described in FIG. 12. [0109] c) Multiplying the
complex envelope by Exp(-j.phi..sub.m(z)) to perform dispersion
compensation.
[0110] Coherent Deconvolution or complex deconvolution for
Dispersion Compensation: Another process known as coherent
deconvolution is explained in [Kulkarni 1999]. One of the inventors
has invented coherent deconvolution methods to correct for imaging
artifacts in OCT. The coherent deconvolution process described in
Kulkarni (1999) comprises of [0111] a) Measuring the interferogram
by placing a mirror in the sample, computing the complex envelope
m.sub.s(z)=A.sub.m(z)Exp(j.phi..sub.m(z)) (Here z is distance in
depth, A.sub.m is amplitude and .phi..sub.m is phase) for the
interferogram, [0112] b) Computing the Fourier transform of
m.sub.s(z) to obtain M.sub.s(k), where k is spatial frequency,
[0113] c) Computing the complex envelope s(z,x) for each
interferogram measurement for any desired specimen, [0114] d)
Computing the Fourier transform of s(z,x) to obtain S(k,x), [0115]
e) Dividing S(k,x) by M.sub.s(k) to obtain S.sub.1(k,x), [0116] f)
Multiplying S.sub.1(k,x) by a Wiener filter to obtain S.sub.1(k,x)
and [0117] g) Computing inverse Fourier transform to obtain
dispersion corrected sample measurement s.sub.2(z, x).
[0118] In FIG. 13, Doppler processing algorithm for high accuracy
and high precision velocity estimation is described (step
1302).
The data set resulting from the camera can be processed in the
processor 114 by the proposed Doppler algorithm which computes STFT
(short time Fourier transforms) in lateral (x) direction (step
1306).
S ^ ( z , x , f ) = m = - N x / 2 N x / 2 - 1 s ( z , ( x + m ) T )
exp [ - j2.pi. fmT ] ( Eq 8 ) ##EQU00009##
where N.sub.x is the number of A-scans in the STFT window. Next the
peak of the STFT spectrum is estimated (step 1308). Next, the
Doppler shift is computed by adaptive centroid algorithm (which
computes centroid using the power near the peak of the STFT
spectrum) (step 1310). Next, the velocity is estimated using
Doppler shifts and Velocity images/maps are generated (step 1312).
Step 1314 is the end of Doppler processing. The velocity precision
is given by
V.sub.c.sup.up=c/(2N.sub.xTv.sub.0n.sub.t cos .theta.) (Eq 9)
[0119] Doppler shift algorithm is used for estimating Doppler
shifts by computing centroid of the short time Fourier transform
spectrum using power near the spectral peak, which is an adaptive
centroid algorithm. As we can see, velocity precision is higher
with higher T (A-scan acquisition period). Therefore, in order to
detect micro-flow (.about.100 to 800 microns/s speed) in
capillaries, by way of example but not by limitation, we can choose
an A-scan rate of e.g., 2560 A scans/s. The maximum retinal blood
flow velocities typically range to 1-4 cm/s. By way of example but
not by limitation, higher velocities can be measured by performing
another scan at a much higher speed of 42000 A scans/s. By way of
example but not by limitation, from Eq. 4, choosing N.sub.x between
1 to 30, we can measure velocities as low as 15 mm/s to 0.5 mm/s,
respectively. By way of example but not by limitation, we can scan
retina at 2 different scan rates, viz., 2560 A scans/s and 42000 A
scans/s. By way of example but not by limitation, in the first set,
we can scan 10 concentric circles centered at the optic disc, each
consisting of 100 A-scans, which can be acquired in 4 seconds. By
way of example but not by limitation, the second set would be
acquired at the same locations, 10 concentric circles, each
consisting of 420 A-scans, which can be acquired in 1 s. The
scanning may be performed by the disc of the retina by performing
concentric circles at a variety of speed. Optical delivery unit in
the sample arm creates scan patterns, wherein the scan-pattern
comprises of at least two B-scans, each B-scan having its specific
A-scan rate.
[0120] Thus, we propose scan-patterns comprising of at least two
B-scans wherein the first B-scan's A-scan rate is slower than the
second B-scan rate.
[0121] The scan-pattern can comprise of at least two B-scans, each
B-scan having its specific A-scan rate.
[0122] This Doppler processing step can used to estimate blood flow
velocities for augmenting diagnosis of diabetic retinopathy. By
acquiring B-scans at various locations, this can be used to obtain
a 3-dimensional map of blood flow velocities or blood vessels in
the retina as well as any organ of a human or animal body.
[0123] The method of FIG. 11 is also applicable for an OFDR-OCT
system. In the OFDR-OCT system, the light entering the detector arm
from the beam splitter is incident on the detector and converts to
an interferometric electric current or signal. The tunable light
source produces a light of various frequencies within a specific
bandwidth. This sweeping is performed at a very high speed and the
detector is able to measure the interference signal at each of the
frequencies. Such a high speed measurement produces a spectrum for
further processing (step 1104 in FIG. 11). These spectra are
typically measured at equal intervals of wavelength. Therefore, the
spectra measured by the detector are processed using a re-sampling
algorithm. Thus, the spectra are resampled at equal intervals of
spatial frequency (k-space) (step 1106). There are some specialized
OFDR-OCT systems where the source is able to sweep the bandwidth at
equal intervals of spatial frequency (k-space). In those cases, the
resampling algorithm is not needed. Next the signal is demodulated
to extract its complex envelope (step 1108). The absolute part of
the complex envelope is traditional OFDR-OCT signal. Next, the
dispersion compensation is performed so that the signal has better
depth resolution and higher fidelity (step 1110). Finally, Doppler
processing is performed to obtain velocity images, which has
velocity information within various locations within a specimen
(step 1112).
[0124] What has been described above includes examples of one or
more embodiments. It is, of course, not possible to describe every
conceivable combination of components or methodologies for purposes
of describing the aforementioned embodiments, but one of ordinary
skill in the art may recognize that many further combinations and
permutations of various embodiments are possible. Accordingly, the
described embodiments are intended to embrace all such alterations,
modifications and variations that fall within the spirit and scope
of the appended claims. Furthermore, to the extent that the term
"includes" is used in either the detailed description or the
claims, such term is intended to be inclusive in a manner similar
to the term "comprising" as "comprising" is interpreted when
employed as a transitional word in a claim.
INDUSTRIAL APPLICATIONS
[0125] OCDR-OCT system and apparatus of this instant application is
very useful for diagnosis and management of ophthalmic diseases
such as retinal diseases and glaucoma etc. Instant innovative
OCDR-OCT diagnostic system leverages advancements in cross
technological platforms. This enables us to supply the global
market a low-cost, portable, robust OCDR-OCT imaging tool, which
would be affordable to general physicians, optometrists and other
health personnel.
[0126] This device can also be used for industrial metrology
applications for detecting depth-dependent flow and micron-scale
resolution thicknesses.
[0127] It is to be understood that the embodiments described herein
can be implemented in hardware, software or a combination thereof.
For a hardware implementation, the embodiments (or modules thereof)
can be implemented within one or more application specific
integrated circuits (ASICs), mixed signal circuits, digital signal
processors (DSPs), digital signal processing devices (DSPDs),
programmable logic devices (PLDs), field programmable gate arrays
(FPGAs), processors, graphical processing units (GPU), controllers,
micro-controllers, microprocessors and/or other electronic units
designed to perform the functions described herein, or a
combination thereof.
[0128] When the embodiments (or partial embodiments) are
implemented in software, firmware, middleware or microcode, program
code or code segments, they can be stored in a machine-readable
medium (or a computer-readable medium), such as a storage
component. A code segment can represent a procedure, a function, a
subprogram, a program, a routine, a subroutine, a module, a
software package, a class, or any combination of instructions, data
structures, or program statements. A code segment can be coupled to
another code segment or a hardware circuit by passing and/or
receiving information, data, arguments, parameters, or memory
contents.
* * * * *