Non-mechanical tomography system

Abstract

A spectral domain optical coherence tomography (OCT) system utilizes an electrowetting prism having four isolated electrodes and configured to deflect a light beam according to an actuating control signal applied to the electrodes. The control signal causes the light beam to laterally scan over a sample. The light beam is produced by a broad bandwidth laser and light from the sample and a reference path is provided to a spectrometer for analysis. The light source may be an all-normal-dispersion fiber laser seeding a gain-managed nonlinear fiber amplifier. The electrowetting prism is capable of scanning up to ±6°. The system is low power and operates at kHz speed.

Description

[0001] NON-MECHANICAL TOMOGRAPHY SYSTEM

[0002] BACKGROUND OF THE INVENTION

[0003] This invention was made with government support under grant number N00014-20- 1-2087 awarded by the Office of Naval Research. The government has certain rights in the invention.

[0004] U.S. Pat. App. No. 63 / 718,135, filed 8 November 2024, is incorporated herein by reference.

[0005] FIELD OF THE INVENTION

[0006] The present invention relates to a non-mechanical tomography system. In particular, the present invention relates to nonmechanical spectral-domain optical coherence tomography using an electrowetting beam scanner.

[0007] DISCUSSION OF RELATED ART

[0008] Optical coherence tomography (OCT) is a non-invasive technique for obtaining high resolution volumetric images. In an OCT system, a near-infrared coherent laser source is scanned laterally across a sample and the reflected light is combined with a path-matched reference beam and collected on a detector. In spectral-domain OCT, a broad bandwidth source is used, and the reflected light is combined with the reference path on a spectrometer. Depth information is contained in the wavelengthdependent interference fringes over the collected spectrum. As a result, spectral- domain OCT is the preferred technique for small-scale, high-speed applications. OCT is an essential biomedical imaging modality capable of achieving high- resolution, volumetric images.

[0009] One common application is in-vivo imaging in which the OCT system is attached to an endoscopic probe. These endoscopes can be crucial for guided surgeries or medical diagnostics. As a result, there have been many recent advances in reducing the overall scale and operating voltage of OCT systems. This is commonly accomplished by replacing the lateral scanning element with a lightweight, compact alternative such as a MEMS mirror. Commercial MEMS mirrors offer a lightweight, high speed scanning method, but tend to be expensive and complicated, and have high operating voltages.

[0010] SUMMARY OF THE INVENTION

[0011] It is an object of the present invention to provide apparatus and methods for nonmechanical imaging. Implementations use non-mechanical beam-steering using optofluidic devices based on the electrowetting-on-dielectric principle.

[0012] Devices are based on the electrowetting-on-dielectric effect in which electrostatic forces from an applied voltage shape the surface of a liquid droplet. As an increasing potential is applied between a polar liquid and a conductive substrate separated by a thin dielectric layer, the contact angle of the droplet changes as a function of the applied voltage. This results in an adjustable prism.

[0013] In some embodiments, an electrowetting prism is capable of scanning at ±6° with an applied voltage of 67 V ±20~V. These capabilities are used to perform lateral scanning in a benchtop spectral domain optical coherence tomography system utilizing a 1091 nm, 53 nm bandwidth gain-managed nonlinear fiber amplifier source.

[0014] Some embodiments accomplish this effect by containing two immiscible liquids, e.g. deionized water (DI water) and 1-phenyl-1- cyclohexene (PCH), in a cylindrical glass cavity attached to a base optical window. The cylindrical cavity is coated in four isolated electrodes and a dielectric layer, and the base window is coated in a ground electrode.

[0015] The electrowetting prism is first actuated to a meniscus curvature near flat by actuating all four electrodes with an equivalent 3 kHz sinusoidal voltage signal (67 V RMS corresponding to approximately 90° liquid interface contact angle). A voltage differential is then applied between opposing electrodes, causing the meniscus surface to tilt by an amount corresponding to the difference in actuated contact angles on opposing electrodes.

[0016] Embodiments are compact, e.g. <4mm in diameter, are forward-viewing, and operate at kHz scan speeds. These devices are useful in a forward-viewing endoscopic pOCT (micro OCT) systems. The invention focuses on spectral domain OCT, in which a broad bandwidth laser source is scanned laterally on a sample plane and the reflected light combined with a reference path and detected by a spectrometer, resulting in a high-speed and high-sensitivity three-dimensional imaging. In-vivo OCT systems rely on the use of a lightweight, compact endoscopic probe. The lightweight, nonmechanical, low power, and transmissive nature of the electrowetting scanning element improves the reliability and lowers the overall SWaP (size, weight, and power) of pOCT endoscopes.

[0017] A spectral domain optical coherence tomography (OCT) system has an electrowetting prism having two isolated electrodes and configured to deflect a light beam according to an actuating control signal applied to the electrodes. It has an imaging path and a path-matched reference path, the imaging path including the electrowetting prism and a scanning lens and configured to scan the light beam across a sample. It also includes a light source and a spectrometer. The control signal is synched to the acquisition of the spectrometer. The light source provides a beam to the reference path and the imaging path and the spectrometer collects recombined light from the reference path and the imaging path for analysis.

[0018] In some embodiments the system has at least four isolated electrodes and can affect both the shape of a liquid-liquid interface within electrowetting prism and it tilt.

[0019] The light source may a broad bandwidth laser, such as a super luminescent diode or an all-normal-dispersion (ANDi) fiber laser configured to seed a gain-managed nonlinear (GMN) fiber amplifier, for example having an ytterbium doped fiber.

[0020] In some embodiments, the electrowetting prism is capable of scanning up to ±6°. The system may have a frame rate of at least 147 kHz, an axial resolution of at least 9 pm in tissue, a sensitivity of at least 95 dB, and at least a 250 pm field of view. Some embodiments exceed a 500 pm field of view.

[0021] The spectral resolution of the spectrometer likely within the sub-nanometer range.

[0022] In some embodiments the system is forward looking. It may be configured as an endoscope with the electrowetting prism and the scanning lens sealed in a sealed element configured to be inserted in vivo. The sealed element may be inserted into an artery, for example. The sealed element may have a diameter of under 4mm.

[0023] A method of performing spectral domain OCT uses a device having an electrowetting prism having at least two isolated electrodes which deflects a light beam according to an actuating control signal applied to the electrodes. The device forms an imaging path (including the electrowetting prism and a scanning lens) and a path-matched reference path. The light beam is provided to the reference path and the imaging path, and the light beam scanned across the sample with the electrowetting prism and the scanning lens. Then, the recombined light from the reference path and the imaging path is providing it to a spectrometer which forms an image.

[0024] With two electrodes, the step of scanning performs a 1 -dimensional scan. If four or more electrodes are used, a two-dimensional scan can be performed. The electrodes actuate a meniscus curvature within the electrowetting prism to near flat by actuating all four electrodes, and then apply a voltage differential between two opposing electrodes, causing the meniscus surface to tilt. For scanning, the tilt is adjusted by varying the voltage differential by steps. In some embodiments, a delay between steps of varying the voltage differential is provided to allow the meniscus to settle.

[0025] BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 (Prior Art) is a schematic side view of an electrowetting prism.

[0027] Figure 2 is a plot showing the beam steering angle of the electrowetting prism as a function of applied voltage differential.

[0028] Figure 3 is a plot showing group velocity dispersion (GVD) for an electrowetting prism.

[0029] Figure 4 is a plot showing spatial dispersion for an electrowetting prism.

[0030] Figure 5 is a schematic block diagram of an OCT device.

[0031] Figure 6 is a plot showing the output spectrum from a gain-managed nonlinear (GMN) fiber amplifier. Figure 7 is a plot showing intensity versus depth for a flat reflector.

[0032] Figure 8 is a plot showing field of view as a function of actuation voltage differential.

[0033] Figure 9 is a plot showing sensitivity roll-off as a function of depth in a sample plane.

[0034] Figure 10 is a plot showing scan reconstruction of a flat reflective surface with attenuation added to the imaging arm.

[0035] Figure 11 shows a scan reconstruction of a sample.

[0036] Figure 12A is an isometric view of a conical resolution target. Figure 12B is a schematic side view showing a profile of a portion of the target.

[0037] Figure 13 is an OCT reconstruction of a scan of a portion of the target.

[0038] Figure 14 is a schematic diagram of a zebrafish eye.

[0039] Figure 15 is an OCT reconstruction of a scan of a portion of a zebrafish eye.

[0040] Figure 16A is a side schematic view of an OCT endoscope. Figure 16B is a magnified cutaway view of the endoscope.

[0041] DETAILED DESCRIPTION OF THE INVENTION

[0042] Figure 1 (Prior Art) is a schematic cross-section view of a 4 mm inner diameter, 5 mm tall electrowetting prism 100. Two immiscible liquids, here deionized water (DI water) 120 and 1-phenyl-1 -cyclohexene (PCH) 118 are contained in a glass cylindrical cavity 106. The cavity interior sidewall is functionalized with 4 discrete electrodes 108 (two are shown in this cross-section) coated in a 3 pm thick uniform layer of Parylene HT 112 and a 600 nm thick layer of Cytop 116. Applying a voltage between DI water 120 and the walls of cavity 106 tunes the contact angle of the liquid interface at each electrode 108.

[0043] Device 100 uses two separately fabricated components: a ground electrode window 104, and a sidewall electrode cavity 106 functionalized with four discrete electrodes 108 around the perimeter of the inner and outer walls. Ground electrode window 104 was fabricated by lithographically patterning annular ground ring electrodes 118 onto a 9.75 x 9.75 x 0.5 mm glass chip 104. A layer of Sll-8 3050 photoresist 112 (= 50 m) was then patterned to serve as an insulation layer between the ground electrode 118 and sidewall electrodes 108.

[0044] The sidewall electrode cavity used a 5 mm tall, 4 mm inner diameter cylindrical glass cavity 106. The inner wall of the cavity is uniformly coated in SU-8 3050 photoresist (not shown). Using a novel direct write laser lithography process, the SU-8 was patterned on the inner walls to create the individual electrode separations needed for beam-steering actuation. This technique reduces the gap between neighboring electrodes to below 30 microns. This helps maintain a large useful aperture and minimizes meniscus-induced distortions on the transmitted beam. The patterned sidewall glass cavity 106 was then uniformly coated in 200 nm of indium tin oxide (ITO), serving as the electrode 108 layer. Excess ITO in undesired regions was removed using a liftoff process in 1-methyl-2-pyrrolidone (NMP) and the conductivity was increased by annealing at 300°C. The cavity was then cleaned and sent for Parylene HT deposition (~ 3 pm thickness). Finally, the entire sidewall cavity was uniformly dip-coated in a hydrophobic Cytop layer 114, (10% weight). This layer increased the initial meniscus contact angle to 173°, extending the device tuning range.

[0045] Finally, the sidewall electrode cavity 106 and ground electrode window 104 were epoxy bonded and attached to a custom printed circuit board (PCB, not shown) allowing electrical connection to be made to each electrode 108 with a 5-pin connector. The device was then filled using a micropipette with equal volumes (approximately 35 pL) of DI water 120 and PCH 118 and then sealed by clamping with an optical window 102. The combination of DI water 120 and PCH 118 was chosen because they are closely density matched (Ap«0.01 g / cm3) and have a large index of refraction contrast (An>0.2), increasing the beam scanning range of the device.

[0046] Figure 2 is a plot showing the beam steering angle of the electrowetting prism as a function of applied voltage differential. A beam-steering angle of ±6° was achieved with an applied voltage differential of ±20V. These results were verified in both the x- direction (dots) and the y-direction (diamonds). To perform lateral scanning, the electrowetting prism 100 is first actuated to a meniscus curvature near flat by actuating all four electrodes 108 with an equivalent 3 kHz sinusoidal voltage signal (67 V RMS corresponding to approximately 90° liquid interface contact angle). A voltage differential is then applied between opposing electrodes, causing the meniscus surface to tilt by an amount corresponding to the difference in actuated contact angles on opposing electrodes. For the given liquid combination of DI water 120 and PCH 118, a beam normally incident on electrowetting prism 100 could be scanned by ± 6° with an electrode voltage differential of ± 20 V. The device has a clear aperture of approximately D = 3.48 mm (approximately 87% of the total aperture) for the maximum scan angle of 6°. Beyond this point, deflected light is clipped by the cylindrical sidewalls. This corresponds to a product of clear aperture and scan angle of 0D = 0.365 mm and a number of resolvable spots equal to N = 9D / 1 ,22k ~ 284.

[0047] In an experimental demonstration, electrowetting prism 100 was used to perform point-to-point scanning across the sample field of view. The prism 100 was first actuated to a bias voltage of approximately 68 V, then actuated to the maximum scan angle by applying a voltage differential of +20 V to two opposing electrodes on the device. Before scanning was initiated, the liquid interface of the electrowetting prism was allowed to settle at this point for approximately 1 second. A number of imaging "pixels" was defined (256 - 1024) and the total scanning differential range was divided into discrete steps corresponding to this value. At each scan "pixel" the liquid interface was again allowed to settle for approximately 250 ms (corresponding to ~1 ,5x the settle time of our electrowetting prism), at the end of which a TTL trigger signal (a digital, binary signal) was sent to the spectrometer to acquire. This process allowed acquisition of an interferogram at a set number of pixels across the field of view defined by electrowetting prism 100. The number of discrete steps depends on the voltage and the number of pixels. For +- 20V (40V total) and 1024 pixels, the resulting voltage step size of 40 / 1024 is 0.039V.

[0048] As an alternative, this process could be performed using a continuous scan. In this case, electrowetting prism 100 would be set to scan at a specified continuous rate across the full voltage differential. In contrast to point-to-point scanning, the total scan time would then be defined by the number of pixels desired and acquisition frequency of the spectrometer.

[0049] Figure 3 is a plot showing group velocity dispersion (GVD) for electrowetting prism 100. Figure 4 is a plot showing spatial dispersion for electrowetting prism 100. One concern when using transmissive beam-steering elements in OCT is the introduction of temporal and spatial dispersion to the system. While dispersion can be corrected during OCT reconstruction, minimization of these effects is helpful for obtaining high resolution OCT images.

[0050] Dispersion is introduced from both the polar liquid, DI water 120, and the nonpolar liquid, PCH 118. The index of refraction in both DI water 120 and PCH 118 differs slightly across the source spectrum. This results in the accumulation of a different optical delay between the imaging 528 and reference path 522 in the OCT system (See figure 5). Beam-steering can also contribute to the dispersion of the system, but this effect is small.

[0051] Net accumulation of optical delay was quantified as a function of path length by deriving the group velocity dispersion (GVD) of PCH as shown in Figure 3. For applications particularly sensitive to dispersive effects, the difference in dispersion between the imaging and reference arm can be compensated by including a cuvette of equivalent path lengths of DI water and PCH in the reference arm.

[0052] To calculate spatial dispersion, the degree of dispersion was approximated across two different sources - 100 nm and 50 nm bandwidth - by calculating the refraction angle at each end of the source spectrum. Two separate beams at opposite ends of the source spectrum passed through the center of a 5 mm long electrowetting prism 100. For the 100 nm bandwidth source (thick line), this corresponded to one beam at 1050 nm and one beam at 1150 nm. Similarly, the 50 nm bandwidth source was calculated for one beam at 1075 nm and one beam at 1125 nm. Assuming a maximum meniscus angle of 30°, both beams were propagated through the device using Snell’s law and the specific index of refraction at these wavelengths for deionized water and PCH. Small differences in the refractive index at each wavelength results in a different steering angle and a lateral separation of the two beams as a function of distance from the exit of our device. This separation is shown in Figure 4. The spatial dispersion for the 100 nm and 50 nm bandwidth source is shown in the top curve and the bottom curve, respectively. For a broad-bandwidth OCT source (AA = 100 nm), the spatial separation of the two beams exceeded 5 pm at approximately 3 mm distance. For the case of a 50 nm bandwidth source, the spatial dispersion exceeded 5 pm after approximately 7.5 mm. Depending on the OCT system, this can affect the lateral resolution achievable. However, it’s important to note that this separation was calculated for the maximum angle of electrowetting device 100. At shallower steering angles, dispersion is greatly minimized. Additionally, smaller electrowetting prisms decrease the thickness of PCH and significantly reduce the spatial dispersion present in the system.

[0053] Figure 5 is a schematic block diagram of an example OCT system 500 used to perform spectral-domain OCT with an electrowetting prism 100. This system uses a 1091 nm center wavelength source 502 with a spectral bandwidth (3 dB) of approximately 53 nm. The source beam 504 is split into an imaging arm 528 and a path-matched reference arm 522. For the imaging arm, the incident beam is first directed through the transmissive electrowetting prism 100, then through a 1 :3 relay telescope 530 (f = 50 mm and f = 150mm) to image the scanning pattern onto the back aperture of an aspheric scan lens 534 (Thorlabs C240TMD-B). The scan lens 534 focuses the incident beam onto the sample plane 534. Reflected light from the sample is then recombined with the reference path 522 and sent via circulator 506 to a spectrometer 510 (Wasatch Cobra 1300) to measure the interferogram at each steering angle. The spectrometer 510 acquired 100 frames at each steering angle at a framerate of 147 kHz. The spectral resolution of the spectrometer was 0.25 nm. This OCT system 500 source 502 used a custom all-normal-dispersion (ANDi) fiber laser with a repetition rate of 35 MHz to seed a gain-managed nonlinear (GMN) fiber amplifier. The GMN fiber amplifier allowed generation of broad-bandwidth (53 nm), compressible pulses centered around 1091 nm. A spectrum of this source is shown in Figure 6.

[0054] Figure 6 is a plot showing the output spectrum 504 from a gain-managed nonlinear (GMN) fiber amplifier 502 as described above. This amplifier balances the nonlinear spectral broadening with a longitudinally evolving gain shape in an ytterbium doped fiber to provide a 1091 nm center wavelength, 53 nm bandwidth (3dB) source. The GMN regime relies on the intentional use of the evolving gain spectrum to manage high nonlinear phase-shifts in a stretcher-less amplification systems. This technique utilizes the full gain bandwidth of the YDF amplification system to reach a broad coherent spectrum. In contrast to incoherent supercontinuum sources, the GMN source can reach similar spectral bandwidths while reducing the pulse-to-pulse noise. This source 502 resulted in OCT images with an axial resolution (« 9 pm in tissue) and sensitivity (110 dB) that matches state-of-the art performance seen in spectral domain OCT systems.

[0055] Figure 7 is a plot showing intensity versus depth for a flat reflector. This was used to measure the axial resolution of system 500. Specifically, system 500 had an axial resolution of 12.8 pm, corresponding to approximately 9 pm in tissue. This is comparable to the 9.7 pm axial resolution predicted by the coherence length of source 502.

[0056] To characterize the axial resolution of system 500, source 502 was imaged onto a flat reflection surface and an interferogram was collected. After performing reconstruction, the intensity trace is shown in Figure 7. Reflection from this surface resulted in a single intensity peak corresponding to the path length difference between reference path 522 and imaging path 528. The full-width at half maximum (FWHM) of this peak gave axial resolution of system 500 and was dependent on the coherence length of source 502. For system 500, measured axial resolution was approximately 12.8 pm in free space, corresponding to approximately 9 pm in tissue. This is similar to the theoretically predicted axial resolution of 9.7 pm for source 502 bandwidth.

[0057] Figure 8 is a plot showing field of view achieved by system 500 as a function of actuation voltage differential. This was found by laterally scanning across a reflective ronchi ruling and observing the number of intensity oscillations over time. System 500 can achieve up to ~ 250 pm field of view with an applied voltage differential of ±20 V. To characterize the field of view achievable by electrowetting prism 100 in system 500, system 500 scanned laterally across a reflective ronchi ruling (100 Ip / mm) and reflected light was collected as a function of scan time onto a photodetector. By counting the number of intensity oscillations over the scan, the number of rulings passed over was determined and the field of view of the scan as a function of actuation angle was approximated. This can be increased by adjusting the magnification of system 500. For system 500, a lateral resolution of below 5 pm was verified when scanning across the ronchi ruling, as each reflective ruling line had a width of 5 pm and full reflection contrast was observed in the collected signal.

[0058] To take OCT measurements, the actuation of electrowetting prism 100 was synchronized to the acquisition of high-speed spectrometer 510. In this example, the same PC / DAQ card 516 was used to provide the sinusoidal actuation voltage 518 to electrowetting prism 100 (via amplifier 520) This created a periodic DC pulse train 530 which was delivered directly to the spectrometer’s trigger input. For static electrowetting scanning, each trigger signal was set to occur after the prism 100 settled at a particular scan angle. The spectrometer 510 acquired 100 frames at each scan angle which were then averaged and used in the reconstruction algorithm.

[0059] For the reconstruction algorithm, the averaged interferogram at each scan angle was first normalized to a reference spectrum. A tapered cosine window was used to optimize the measured axial resolution of system 500 while minimizing the effect of reconstruction sidelobes. To find the optimal window size, a pattern-search method used the minimization of both the axial resolution and the DC component of the reconstruction as the optimization parameters. Additionally, chromatic dispersion present in the source 502 spectrum 504 was corrected for by subtracting higher order dispersion terms from the interferogram. Finally, the normalized and windowed interferogram was then Fourier transformed to arrive at the A-scan reconstruction at each scan angle.

[0060] Figure 9 is a plot showing sensitivity roll-off of system 500 as a function of depth in the sample plane 534. This roll-off is dependent on the sampling rate of spectrometer 510. This was measured by taking successive A-scan reconstructions at various path delays and overlaying each. The sensitivity of system 500 begins to roll-off around 1.5 mm.

[0061] The imaging depth achievable from system 500 is limited by the sampling rate of spectrometer 510 (Wasatch Cobra 1300). For reflection points that are far from path- matched, the frequency of the fringes collected on an interferogram exceeded the Nyquist sampling rate and became aliased. For spectrometer 510, this was predicted to be approximately 1 .47 mm. To measure the depth limit of system 500, an interferogram from a flat reflector was recorded at a range of reference arm 522 delay lengths while electrowetting prism 100 was actuated at 67 V (90°contact angle). An A-scan reconstruction was performed at each position and these results overlayed as shown in Figure 9. Sensitivity of system 500 began to roll-off at ~ 1 .5 mm.

[0062] Figure 10 is a plot showing scan reconstruction of a flat reflective surface with 70 dB attenuation added to the imaging arm. This was done to determine the noise floor of system 500. The maximum sensitivity of system 500 is approximated by taking the height of the reconstruction peak above the noise floor and adding back in the 70 dB of attenuation. System 500 had a maximum sensitivity of approximately 110 dB.

[0063] To measure the overall sensitivity of system 500, an interferogram from a flat reflector was recorded when electrowetting prism 100 was actuated at 67 V (90° contact angle). The signal in the imaging arm was then attenuated until interference fringes measured on spectrometer 528 were barely visible. This measurement was used as the minimum sensitivity above the noise floor. The sensitivity of the reconstructed peak in this measurement was adjusted for the 70 dB of attenuation to find peak intensity. System 500 has a sensitivity of approximately 110 dB, comparable to previous spectral domain OCT demonstrations. It is important to note that this sensitivity was acquired when device 100 was actuated to flat. Spatial and temporal dispersion represent the largest contributions to degraded imaging quality at high deflection angles. However, since electrowetting device 100 is actuated to relatively low scan angles (< 6°), this effect is likely to be small. Specifically, the optical pathlength difference between flat and the maximum steering angle of < 6° was approximately 13.8 pm for a 5 mm long device. Using the value for the GVD of PCH, this is equivalent to a small 1.58 fs2difference in Group Delay Dispersion (GDD). Similarly, as shown in Figure 3, spatial dispersion for a 50 nm bandwidth source at the maximum steering angle exceeds the lateral resolution (5 pm) after approximately 7.5 mm.

[0064] To demonstrate the ability of electrowetting prism 100 to perform successful OCT reconstructions, images were taken of multiple test samples including a layer of diamond polishing paste (Figure 11 ), a 3D printed conical step target (Figure 13), and zebrafish eye (Figure 15). Each image was acquired by first actuating electrowetting prism 100 to flat (90° contact angle corresponding to 67 V) and applying a voltage differential to scan laterally across the sample of interest.

[0065] Figure 11 shows the B-scan reconstruction of a 175 pm thick layer of diamond polishing paste 1102 containing individual diamond beads1104 with diameters ranging from 10 - 20 pm in diameter. The magnified view AA shows that system 500 can resolve individual diamond beads 1104 throughout the depth of the sample 534.

[0066] This image was taken by actuating electrowetting prism 100 at ± 20 V voltage differential. A total of 400 individual averaged interferograms were collected over the entirety of the scan.

[0067] Figure 12A is an isometric view of a conical resolution target 1202. Figure 12B is a schematic side view showing a profile of a portion of the target 1202, showing steps 1206. A custom 3D-printed conical step target 1202 was imaged by incident light 1204. Each step 1206 on the target was about 75 pm in height and width. As source 502 was scanned laterally across a section of target 1202, each step 1206 was resolved as a separate reflecting surface separated by about 75 pm.

[0068] Figure 13 shows a B-scan OCT reconstruction obtained of sample 1202. Prism 100 was actuated with a differential voltage of ±20 V, corresponding to a lateral field of view of approximately 250 pm. System 500 resolved three distinct steps across the imaging field of view. The tilt and slight variation in the step height present in the reconstruction of Figure 13 was a result of 3D printing errors, not reconstruction artifacts. This was confirmed by taking a profile measurement on a standard optical profilometer (Keyence VKX).

[0069] Figure 14 is a schematic diagram of a zebrafish eye 1400 having lens 1402, cornea 1404, and iris 1406. Figure 15 is an OCT reconstruction of a scan of a portion of the zebrafish eye 1400. The zebrafish eye is an important analogue for human ocular genetics and serves as a useful proof-of-concept for biological tissue imaging using system 500. System 500 was focused onto the top surface of the eye to identify large scale ocular structure. Figure 15 shows the cornea 1404 and iris 1406 of the sample. Electrowetting prism 100 was actuated at a voltage differential of ± 20 V, corresponding to a lateral field of view of approximately 250 pm.

[0070] Figure 16A is a schematic side view of an OCT endoscope 1600 based on system 500, disposed within an artery 1610 and connected to optical cable 1602. It shows an example of how the device is assembled, its overall dimensions, and how it is used. Figure 16B shows a cutaway view of endoscope 1600, comprising a coupling lens 1604, electrowetting scanner 100, and scan lens 1608, all sealed in a housing about 15mm long. A 3mm diameter optical cable 1602 couples to endoscope 1600. Note that the reference path 522 is not depicted in Figure 16. It is on a separate board (not shown) connected to endoscope fiber 1602.

[0071] Endoscopes benefit from reducing the voltage of the device. An electrowetting prism capable of scanning at less than +-10V can be achieved by using a thinner dielectric coating, an angled sidewall geometry, and a lower initial contact angle (e.g. using different superhydrophobic coating).

[0072] While the exemplary preferred embodiments of the present invention are described herein with particularity, those skilled in the art will appreciate various changes, additions, and applications other than those specifically mentioned, which are within the spirit of this invention

[0073] What is claimed is:

Claims

CLAIMS1 . A spectral domain optical coherence tomography (OCT) system comprising: an electrowetting prism having two isolated electrodes and configured to deflect a light beam according to an actuating control signal applied to the electrodes; an imaging path and a path-matched reference path, the imaging path including the electrowetting prism and a scanning lens and configured to scan the light beam across a sample; a light source; a spectrometer; and wherein the control signal is synched to the acquisition of the spectrometer; wherein the light source provides a beam to the reference path and the imaging path; and wherein the spectrometer collects recombined light from the reference path and the imaging path for analysis.

2. The system of claim 1 including four isolated electrodes and configured to affect both the shape of a liquid-liquid interface within the electrowetting prism and the tilt of the liquid-liquid interface.

3. The system of claim 2 wherein the light source is a broad bandwidth laser.

4. The system of claim 3 wherein the light source has center wavelength within 10% of 1091 nm with a spectral bandwidth (3 dB) of 50-80 nm.

5. The system of claim 3 wherein the source is an all-normal-dispersion (ANDi) fiber laser configured to seed a gain-managed nonlinear (GMN) fiber amplifier.

6. The system of claim 5 wherein the ANDi fiber laser comprises an ytterbium doped fiber.

7. The system of claim 3 wherein the source is a super luminescent diode.

8. The system of claim 2 wherein the electrowetting prism is capable of scanning up to ±6°.

9. The system of claim 2 configured to have a frame rate of at least 147 kHz.

10. The system of claim 2 wherein the spectral resolution of the spectrometer is configured to be in the sub-nanometer range.11 . The system of claim 2 configured to have an axial resolution of at least 9 pm in tissue12. The system of claim 2 configured to have a sensitivity of at least 95 dB.

13. The system of claim 2 configured to have at least a 250 pm field of view.

14. The system of claim 13 configured to have at least a 500 pm field of view.

15. The system of claim 2 configured to be forward looking.

16. The system of claim 2 configured as an endoscope with the electrowetting prism and the scanning lens sealed in a sealed element, the sealed element configured to be inserted in vivos.

17. The system of claim 16 wherein the sealed element is configured to be inserted into an artery.

18. The system of claim 16 wherein the sealed element has a diameter of under 4 mm.

19. The method of performing spectral domain OCT comprising the steps of: providing a device having an electrowetting prism having two isolated electrodes and configured to deflect a light beam according to an actuating control signal applied to the electrodes and forming an imaging path and a path-matched reference path, the imaging path including the electrowetting prism and a scanning lens and configured to scan the light beam across a sample; providing a light beam to the reference path and the imaging path; scanning the light beam across the sample;collecting the recombined light from the reference path and the imaging path and providing it to a spectrometer; and analyzing the recombined light to form an image of the sample.

20. The method of claim 19 wherein the step of scanning performs a 1- dimensional scan.21 . The method of claim 19 wherein the electrowetting prism includes four isolated electrodes and the step of scanning the light beam across the sample further comprises the steps of: actuating a meniscus curvature within the electrowetting prism to near flat by actuating all four electrodes; applying a voltage differential between two opposing electrodes, causing the meniscus surface to tilt; and varying the voltage differential by steps to adjust the tilt.

22. The method of claim 21 further including the step of providing a delay between steps of varying the voltage differential to allow the meniscus to settle.

Images