Line field OCT system with k-space calibration

Abstract

Line-field optical coherence tomography (OCT) systems and methods provide enhanced imaging accuracy through k-linearization of interferometric data. The system comprises a swept-frequency laser source, a line-field sensor, and a single-board computer (SBC) with processing capabilities. The system uses a frequency reference to generate a periodic reference pattern detected by a subset of pixels in the line-field sensor. The SBC determines a resampling curve based on the reference pattern to correct for nonlinearities in the laser's wavelength tuning. Applying the resampling curve to k-linearize the interferometric data enables the generation of high-resolution transform-limited depth profiles via inverse Fourier transform. Methods for periodically, continuously, or adaptively recalculating the resampling curve based on a linearity threshold are disclosed. The system can also dynamically update the laser's tuning function to ensure consistent performance. These advancements enable accurate and efficient OCT imaging for applications such as ophthalmology, angiography, and other diagnostic uses.

Description

[0001] Related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 609,916, filed December 14, 2023, pursuant to 35 USC §119(e), which is incorporated herein by reference in its entirety. Background Technology

[0003] Optical coherence tomography (OCT) is a cross-sectional, non-invasive imaging technique used in many areas of medical imaging. For example, in ophthalmology, OCT has been widely used to image the retina, choroid, and anterior segment. Functional imaging of blood flow velocity and microvessels is also possible.

[0004] Fourier domain OCT (FD-OCT) has recently gained more attention due to its higher sensitivity and imaging speed compared to time domain OCT (TD-OCT), which uses optical delay lines for mechanical depth scanning, resulting in a relatively slow imaging speed. The spectral information differentiation in FD-OCT is achieved by using a dispersive spectrometer (spectral domain or SD-OCT) or a fast-scanning swept-frequency laser source (swept-frequency source OCT or SS-OCT) in the detection arm.

[0005] Compared to spectrometer-based FD-OCT, SS-OCT has several advantages, including its robustness against motion artifacts and stripe erosion, lower sensitivity roll-off, and higher detection efficiency.

[0006] Commercial ophthalmic scan-source OCT systems are traditionally flying-spot systems. This system scans the eye and resolves the A-scan at each point. Volume is typically built up by scanning points across a raster across the eye (usually the retina). Raster scanning is conventionally achieved using a pair of orthogonal galvanometers. This translates to a very challenging requirement for the scan source, which must scan its scan band for each point before moving to the next.

[0007] Many different methods have been implemented to develop high-speed swept laser sources for SS-OCT. One approach is to employ a ring laser design based on a semiconductor optical amplifier (SOA) (see, for example, Yun et al., “High-speed optical frequency-domain imaging”, Opt. Express 11:2953 2003 and Huber et al., “Buffered Fourier domain mode locking: unidirectional swept laser sources for optical coherence tomography imaging at 370,000 lines / s”, Opt. Express 13,3513 2005). Another example is short-cavity lasers (see, for example, "Compact Ultrafast Reflective Fabry-Perot Tunable Lasers For OCT Imaging Applications" by Kuznetsov et al., Proc. SPIE 7554:75541F 2010). SOA-based ring laser designs are practically limited to positive wavelength sweep (increasing the wavelength) because significant power losses occur in negative tuning. This is because four-wave mixing (FWM) in the SOA causes a negative frequency shift in the intracavity light as it propagates through the SOA (Bilenca et al., “Numerical study of wavelength-swept semiconductor ring lasers: the role of refractive-index nonlinearities in semiconductor optical amplifiers and implications for biomedical imaging applications”, Opt. Lett. 31: 760-762, 2006).

[0008] Meanwhile, other architectures exist for SS-OCT that reduce the performance requirements of the swept laser source. Fechtig et al., in their article titled "Line-Field parallel swept source MHz OCT for structural and functional retinal imaging," *BiomedicalOptics Express* 716, Vol. 6, No. 3 (2015), described a system that achieves an equivalent A-scan rate of 1 MHz by combining a lower sweep rate laser with a linear line-field sensor. Even earlier examples exist, such as Lee et al.'s "Line-Field Optical Coherence Tomography Using Frequency-Sweeping Source," published in *IEEE Journal of Selected Topics in Quantum Electronics*, Vol. 14, No. 1, January 2008.

[0009] Interferometric data collected by the SS-OCT system needs to be uniformly distributed in wavenumber space (k-space) because the Fourier transform used to reconstruct the depth profile (A-scan) assumes that the sampled data is uniformly distributed in k-space. This assumption is crucial for achieving the highest transform-limited resolution along the axial (depth) direction.

[0010] If the spectral data is not uniformly distributed in the wavenumber space k-space, nonlinear sampling introduces distortion into the Fourier transform, leading to artifacts in the resulting image, reduced axial resolution, and inaccurate depth measurements. This is because the Fourier transform inherently relies on the uniformity of the sampling interval to correctly resolve spatial frequencies. Non-uniform sampling causes aliasing or misrepresentation of depth structures, thus degrading image quality.

[0011] K-linearization corrects for this by resampling the spectral data to align it with points uniformly spaced in k-space, thus ensuring the Fourier transform operates as expected. This step is particularly important for applications requiring high-resolution imaging, such as ophthalmic diagnostics or microvascular imaging, where precise axial detail is crucial.

[0012] Most sweep sources used in SS-OCT systems are controlled to be linearly tuned across the scan band. However, most are not tuned linearly enough to achieve the highest possible performance. K-linearization refers to the process of resampling the acquired spectral interferometric data to make it uniformly distributed in k-space.

[0013] To achieve k-linearization, a frequency reference (such as an etalon) is used to monitor the instantaneous wavenumber during laser scanning. Detected spectral fringes from the reference are analyzed to determine deviations from the ideal linear tuning. A resampling curve is then calculated to uniformly redistribute the spectral data points in k-space. This ensures that the interferometric data corresponds to uniformly spaced wavenumbers, thereby improving the quality of depth profiles and tomographic images. K-linearization is critical in applications requiring high axial resolution and is typically implemented as a preprocessing step prior to Fourier transform. Summary of the Invention

[0014] For the first order, K-linearization can also be achieved by carefully controlling the tuning characteristics of the sweep source. For common operating wavelengths, the sweep source must sweep through a relatively wide scan band. For systems operating in the 800-900 nm band, the sweep source should sweep through a scan band greater than 50 nm. Systems operating at even longer wavelengths will have scan bands approaching 100 nm wide. Tuning such a relatively wide band of laser requires some type of intracavity mechanical tuning system to tune the laser. In some examples, micromechanical systems (MEMS) films and Fabry-Perot tunable filters are used. Other methods include tilted bandpass interferometer filters and tilted gratings. In all these cases, there is a transfer function between the intracavity mechanical tuning system and the instantaneous wavelength of the laser. K-linearization requires understanding the transfer function so that the laser is linear in frequency through the sweep of the scan band, or at least as linear as possible given mechanical constraints.

[0015] Both resampling and k-linearization require a frequency reference across the sweep period to accurately monitor the instantaneous frequency of the laser. In conventional systems, this frequency reference is often obtained using a separate fiber interferometer or etalon that receives a portion of the laser's emission, and the transmission of the interferometer or etalon or similar spectral reference is monitored using a separate detector and analog-to-digital converter.

[0016] This invention utilizes one or more pixels of a line field sensor to track sweep linearity and thus enables k-linearization.

[0017] Generally, according to one aspect, the present invention features an optical coherence tomography (OCT) system comprising a swept laser source emitting light having wavelengths tuned across a predefined scan band. A line field sensor is arranged to receive an interference signal corresponding to light scattered from a sample and a reference light, while detecting a periodic reference pattern generated by the interaction of a frequency reference with a portion of the output of the swept laser source. By dedicating a subset of a linear array of the sensor's pixels (e.g., those pixels at one end of the array) to capturing these reference signals, the system obtains a known periodic pattern reflecting the instantaneous wavenumber of the laser. Using this reference pattern, a computer or processor determines a resampling curve to compensate for any nonlinearity in the wavelength tuning of the swept laser source. The resampling curve can be applied to the interference signal to achieve k-linearization, and subsequently, an inverse Fourier transform is performed on the k-linearized data to produce depth profiles. These profiles are combined to form a tomographic image displayed to a user. In some embodiments, the system includes a scanning mechanism to translate the illumination line over the sample, thereby generating a series of k-linearized depth profiles at different lateral positions for full-volume imaging.

[0018] In other aspects, the swept laser source is equipped with an intracavity mechanical tuning mechanism and a tuning profile module that stores a predefined tuning function that governs the wavelength sweep. The controller evaluates the tuning linearity based on a detected reference signal and determines a resampling profile to correct for any observed nonlinearities. Furthermore, the controller can update the predefined tuning function in the tuning profile module to achieve a more consistent wavelength sweep in future scans. The result is an OCT system that can adaptively recalibrate both software-based k-linearization and hardware-based tuning functions over time.

[0019] Further embodiments involve using an etalon or other frequency reference element that covers only a portion of the pixel array of the line field sensor. By limiting the detection of the reference pattern to a known set of “reference pixels,” the system can simultaneously acquire both interferometric sample data and periodic patterns for k-calibration. The reference pixels provide the necessary instantaneous wavenumber information, thereby ensuring that the system can compensate for frequency nonlinearities as needed, either in real-time or periodically, to maintain stable and accurate axial resolution.

[0020] Methods for OCT imaging involve capturing interferometric and reference signals during the tuning of a swept laser source, determining a resampling curve from the reference pattern to address tuning nonlinearities, and applying this curve to k-linearize the interferometric data. Following k-linearization, an inverse Fourier transform reveals a depth profile suitable for tomographic imaging. Optionally, the method includes updating a predefined tuning function for the swept source based on the determined resampling curve, thereby improving the linearity of future wavelength sweeps.

[0021] In another method for calibrating a frequency-sweeping laser source, a reference signal generated from a frequency reference is captured and analyzed to assess wavelength sweep linearity. From these measurements, resampling curves and / or updates to the stored tuning function are derived, enabling the system to ensure that the acquired interferometric signal is k-linearized. Therefore, this calibration method lays the foundation for generating high-quality tomographic images with improved axial resolution and reduced artifacts.

[0022] These embodiments and methods enhance OCT system performance by directly integrating a frequency reference in front of a subset of pixels of the online field sensor, enabling simultaneous acquisition of sample and reference data. The resulting dynamic adjustment of both the resampling curve and the predefined tuning function produces more accurate, robust, and stable OCT imaging.

[0023] The above and other features of the invention, including various new construction details and combinations of components, as well as other advantages, will now be described in more detail with reference to the accompanying drawings, which are set forth in the claims. It will be understood that the specific methods and apparatus for carrying out the invention are shown by way of illustration and not by way of limitation. The principles and features of the invention may be employed in various and many embodiments without departing from the scope of the invention. Attached Figure Description

[0024] In the accompanying drawings, reference numerals are used throughout the different views to denote the same parts. The drawings are not necessarily drawn to scale; rather, the focus is on illustrating the principles of the invention. In the accompanying drawings:

[0025] Figure 1 This is a schematic diagram of the cat's eye tunable laser in a line field parallel sweep frequency source OCT system;

[0026] Figure 2 It is a diagram of the tuning curve and angle of the laser frequency and angle control actuator 132, as well as the periodic transmission of the etalon that provides a frequency or wavenumber reference during the collection of one thousand line interference signals used for B-scan.

[0027] Figure 3 This is a schematic top view showing a pixel array of a line field sensor with a periodic frequency reference; and

[0028] Figure 4A , Figure 4B , Figure 4C and Figure 4D This is a flowchart illustrating different methods for calculating new resampling curves and / or tuning curves for k-linearization. Detailed Implementation

[0029] The invention will now be described more fully below with reference to the accompanying drawings, which illustrate illustrative embodiments of the invention. However, the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to make this disclosure thorough and complete, and to fully convey the scope of the invention to those skilled in the art.

[0030] As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items. Furthermore, all conjunctions used should be understood in the most inclusive sense possible. Therefore, unless the context explicitly requires it, the word “or” should be understood as having the definition of a logical “or” rather than a logical “exclusive or”. Additionally, unless explicitly stated otherwise, the singular form and the articles “a” and “the” are also intended to include the plural form. It will be further understood that, when used in this specification, the terms include, comprise, including, and / or comprising specify the presence of the stated feature, integer, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Additionally, it will be understood that when an element (including components or subsystems) is mentioned and / or shown as connected to or coupled to another element, it may be directly connected to or coupled to that other element, or there may be intermediate elements present.

[0031] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It will also be understood that terms (such as those defined in common dictionaries) shall be interpreted as having the same meaning as they have in the context of the relevant field, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0032] Figure 1 An OCT system 200 with a line field or line parallel sweep frequency source is shown. In a preferred embodiment, a cat's eye tunable laser 100 is used as the sweep frequency source, but the invention can be applied to other tunable laser technologies, such as lasers tunable by MEMS, gratings, acousto-optic tunable filters, to name a few examples.

[0033] The present invention is also applicable to sweep source OCT systems 200 that achieve k-linearity by resampling instead of relying on controlling the sweep source to be linearly tuned in frequency, or to those systems that employ a combination of these two methods.

[0034] In one example, laser amplification is provided by a GaAlAs gain chip 110. The gain chip 110 amplifies light in a wavelength range of approximately 800 nm to 900 nm. Its center wavelength is preferably approximately 840 nm, which is useful for applications such as ophthalmic imaging and other diagnostic uses because of the water window (650 nm to 950 nm) at these wavelengths. Another advantage of this wavelength range is that it can be detected using a standard camera with a silicon-based imager chip or sensor. Specifically, the output is detected by a silicon imager (e.g., complementary metal-oxide-semiconductor (CMOS) or charge-coupled device (CCD)).

[0035] In other examples, the lasers operate within different wavelength ranges. Other examples operate around 1050 nm or 1310 nm, which are locations of other water windows. However, the problem with these ranges is that they require InGaAs linefield sensors that are more expensive than commercial silicon devices operating at shorter wavelengths.

[0036] In the current embodiment, the gain chip 110 is mounted in a TO-can type sealed package 112. This protects the chip 110 from dust and the surrounding environment, including moisture. In some examples, the TO-can package has an integrated or separate thermoelectric cooler 114.

[0037] Chip 110 is preferably a single-angle endface (SAF) edge-emitting chip. Thus, it has a high reflectivity (HR) coated rear endface 150. It has an anti-reflection (AR) coated front endface 152. Furthermore, for improved performance, it has a curved ridge waveguide 154 that is perpendicular to the rear endface 150 but angled at the interface with the front endface 152. This angle at the front endface, along with the AR coating, reduces reflections at the front end reflectivity by up to 40 dB and significantly improves laser performance by reducing parasitic reflections that would otherwise lead to unsmooth tuning and mode hopping.

[0038] The free-space beam 116 from package 112 diverges along two axes (x, y). It is collimated by collimating lens 118. The resulting collimated beam 124 is received by cat-eye focusing lens 120, which focuses the light onto cat-eye / output coupler 122. This defines the other end of the laser cavity extending between the back / reflection facet of gain chip 110 and the output / reflection facet of output coupler 122.

[0039] Collimated light 124 between collimating lens 118 and cat's-eye focusing lens 120 passes through thin-film interference bandpass filter 130. This provides a passband with approximately 0.3 nanometers (nm) full width at half maximum (FWHM) for an OCT application in a specific example. More generally, its passband is often between 0.2 nm and 0.5 nm FWHM, or more generally between 0.1 nm and 2 nm FWHM. Even more generally, it is between 0.05 nm and 5 nm FWHM.

[0040] A bandpass filter is held on the arm of an angle control actuator 132, which changes the angle between the bandpass filter 130 and the collimated beam 124. In one example, the angle control actuator is a galvanometer. In other examples, the angle control actuator 132 is a servo motor or electric motor that causes the bandpass filter 130 to rotate continuously within the collimated beam 124. This allows the bandpass filter 130 to tilt relative to the collimated beam 124, thereby tilting the tuned filter and thus changing the passband to scan or sweep the wavelength of the frequency-sweeping laser 100.

[0041] The tuning speed specifications for galvanometers typically range from 0.1 Hz to 50 kHz. For higher speeds, resonant galvanometers may be used with bidirectional tuning, but both higher and lower speeds are possible. Wavelength tuning speeds are usually given in nm / sec, so for an ideal tuning speed of 100 Hz for retinal imaging applications, a linear velocity camera capturing frames at 100 kHz would give 1000 sampled bandwidth points and a 70 nm tuning range, resulting in 70 nm / 10 msec = 7000 nm / sec. Generally, tuning speeds should be between 3000 nm / sec and 11000 nm / sec or higher.

[0042] For retinal or industrial imaging using low-cost CMOS or CCD cameras, a center wavelength of 840 nm is an ideal water window. The tuning range is typically at least 30 nm. Preferably, the tuning range is closer to 60 nm or 70 nm or larger. This provides good resolution of <8 micrometers in air. Generally, the tuning range should be between 30 nm and 100 nm.

[0043] The size of the collimated beam 124 is important for many applications. As a general rule, a smaller beam results in higher divergence, which in turn results in a larger cone half-angle (CHA). This reduces the minimum linewidth with respect to angle for tunable filters. In the current embodiment, the collimated beam is preferably not less than, i.e., greater than 1 mm FWHM, and for retinal OCT applications, preferably greater than 2 mm FWHM. Generally, the CHA should be less than 0.04 x 0.02 degrees, and preferably about 0.02 x 0.01 degrees or less.

[0044] The light from the gain chip is polarized. In common architectures, the polarization is horizontal or parallel to the epitaxial layer of the edge-emitting gain chip 110. In a preferred configuration, the filter is oriented to receive S-polarization in order to maintain a narrow linewidth when it is tilted during tuning. On the other hand, P-polarization broadens sharply at large tilt angles. S-polarization has higher loss than P-polarization at large tilt angles. Therefore, in the current embodiment, the filter design needs to address these issues by providing sufficiently low loss across the S-polarization tuning band.

[0045] On the other hand, in some embodiments, a P-polarization configuration may be desirable due to the higher power across the scan band.

[0046] Generally, current cat's eye configurations offer several advantages. They provide low-loss, low-tolerance, and repeatable stable operation because they offer lower angular wavelength variations than grating-based lasers and can be manufactured at a lower cost than AOTF and MEMS lasers.

[0047] The mirror / output coupler 122 typically reflects approximately 80% of the light back into the laser cavity and transmits approximately 20%. More generally, the mirror / output coupler can reflect 10% to 99% of the light (transmitting 90% to 1%, respectively), depending on the desired output power and laser cavity losses. Higher reflectivity results in lower cavity losses and therefore a wider laser tuning range where gain exceeds loss, but leads to lower output power. In typical operation, the mirror / output coupler 122 reflects less than 90%.

[0048] In some embodiments, an aperture or shield 190 is typically added after the mirror output coupler 122 to trim the beam edge. This reduces power fluctuations caused by beam drift due to refraction in the tilted bandpass filter 130. Preferably, it is between 80% and 95% of the beam size, and more preferably approximately 90% of the beam size.

[0049] Typically, the diverging beam from the mirror output coupler 122 is collimated with the output collimating lens 140 to form a free-space output beam 102.

[0050] The angle control actuator 132 operates as a servo mechanism. In the illustrated embodiment, the angle control actuator 132 is a servo-controlled galvanometer with an encoder 160. The encoder 160 generates an angle signal 162 that indicates the angle between the galvanometer and, consequently, the filter 130 and the collimated beam 124. Preferably, the encoder is an optical encoder and is often analog.

[0051] The laser controller 232 receives the angle signal 162 at the PID (Proportional-Integral-Derivative) controller 164. The PID controller 164 compares the angle signal 164 with a specified tuning function stored in the tuning curve module 166. The desired tuning curve is often stored in a lookup table or generated algorithmically. This tuning curve is defined such that the laser 100 will linearly sweep its scan band in frequency, thereby providing or contributing to k-linearization. This is often an approximately sawtooth or triangular waveform. The PID controller 164 generates a control function 168, which is used to drive the windings of the galvanometer 132 via the amplifier 169.

[0052] In the example shown, the OCT system 200 is used to analyze a human eye 202, specifically the retina 204. That is, the system can also be used to analyze other samples, including both biological and non-biological samples.

[0053] In the example shown, light from laser 100 in the form of a free-space beam 102 is transmitted in free space to line-forming optics 208, and then to beam splitter 210 of an OCT interferometer, such as a cube beam splitter.

[0054] Typically, the line-forming optics 208 includes one or more cylindrical lenses and may include several additional lenses in a beam expander configuration. The line-forming optics 208 converts light from the laser 100 into a line, or more specifically a rectangular cross-section, with an aspect ratio of at least 10:1 and typically greater than 100:1, and often 400:1 or greater. That is, when viewed along its optical axis, light from the line-forming optics 208 has a line, or more specifically a rectangular two-dimensional cross-section, which is at least 10 times longer in one dimension than in another, as measured by FWHM.

[0055] Beam splitter 210 divides light between reference arm 212 and sample arm 214 in the illustrated Michelson arrangement. Light propagates in the free space between one or more lenses that form projections and collect optics 222 in the sample arm and illuminates sample 202, typically tissue 204, such as the retina or cornea of ​​the human eye.

[0056] The light rays are scanned by a line scanner 280. In some examples, the scan is radial. In other examples, the line is scanned as a brush, that is, in a direction orthogonal to its principal axis.

[0057] On the other hand, the light in the reference arm 212 is modulated by one or more lenses of the reference arm optics 224 and reflected by the reference mirror 226.

[0058] The collected sample light, received by the projection and collection optics 222, is combined with the reference arm light to form an optical interference in the line field camera or sensor 230. Line field sensors typically have a linear array of at least 512 pixels, and often at least 1024 or 2048 pixels, to detect interference signals for lines. In the current example, the linear array is a few pixels wide, such as between 2 and 10 pixels wide. Pixels can often be binned along this horizontal axis to obtain higher sensitivity. Multiple vertically arranged linear arrays can also be binned to give an effective higher pixel count and result in improved sensitivity.

[0059] The output from sensor 230 is read by a single-board computer 235, which also controls system 200 and specifically controls laser controller 232. In this example, single-board computer 235 is a modular system (SOM) that includes a graphics processing unit (GPU), a central processing unit (CPU), memory, power management, and high-speed interfaces. Currently, a Jetson Orin series module from NVIDIA is used. Alternatively, signals can be read from the sensor using a PC via USB, Gigabit Ethernet (GigE), coaxpress, or via a frame capture device.

[0060] The output from the line field sensor 230 is stored in the SOM 230 and then reconstructed for display on the display 234. The Fourier transform of the interfering light is performed by the GPU within the SOM 230. This reveals a profile of the scattering intensity at different path lengths, thus revealing the scattering according to depth (z-direction) in the sample (see, for example, “Ultrahigh resolution Fourier domain optical coherence tomography” by Leitgeb et al., Optics Express 12(10):2156 2004). The profile of scattering according to depth at a point is called the axial scan (A-scan). The combination of the projection line and the line scan sensor 230 produces a cross-sectional image of the sample (tomogram or B-scan).

[0061] In one implementation, the SOM 235 runs OCTproZ, an open-source software for optical coherence tomography (OCT) processing and visualization, available at github.com / spectralcode / OCTproZ. A plug-in system enables the integration of custom OCT systems and software modules.

[0062] Figure 2 It is a graph showing the laser frequency or wavenumber used to collect the line interference signal required for B-scan, and the angle of the angle control actuator 132.

[0063] More specifically, control logic 262 triggers the start of the scan using trigger 260. The laser 100 is swept along a wavelength such that its frequency preferably changes linearly with time during the sweep or B-scan period. Linear frequency sweeping is often desirable because it allows for efficient use of the camera's sampling rate and, in some embodiments, even eliminates the need for resampling.

[0064] For a linear frequency sweep, the angle of filter 130 must be tuned in a non-linear manner. As shown, the rate of change of the angle of filter 130 slows down with increasing angles, shorter wavelengths, and higher frequencies of laser emission 102. In one example, the tuning profile for the linear frequency sweep is generated and / or stored in tuning profile module 166 and is updated by control of SOM 235 by downloading new lookup tables or tuning profile algorithms into tuning profile module 166.

[0065] When laser 100 begins its sweep, control logic 262 of laser controller 232 generates trigger signal 260, which initiates the capture of line interference signals during laser tuning. Preferably, at least 250 line interference signals are captured by line scan camera 230 during the sweep period of laser 100. Currently, more than 500 line interference signals, such as 1000 or more, are captured.

[0066] Preferably, the sweep period is less than 0.05 seconds, and more preferably less than 0.02 seconds. In the current example, a sweep period of approximately 0.01 seconds or less provides an acceptable B-scan of the human eye, despite the presence of microsaccades and other movements.

[0067] In the current implementation, the line field camera 230 operates at approximately 100 k LPS (lines per second) for a sweep period of approximately 0.01 seconds. Generally, the line field camera 230 should operate at greater than 20 k LPS to ensure high-quality B-scans despite microsaccades, and preferably at greater than 100 k LPS. In fact, it can operate at greater than 100 k LPS when performing OCT angiography.

[0068] Several B-scans (typically more than 10) have often been acquired and averaged. Control logic 262 controls scanner 280 to move the projection lines on the patient's retina. The process is repeated, capturing several lines and generating averaged B-scans for new positions.

[0069] The SOM 235 performs an inverse Fourier transform to obtain a depth profile from the acquired stripe pattern. Several preprocessing steps are often performed before the IFFT, including background removal, k-linearization, dispersion compensation, and / or windowing.

[0070] K-linearization should be optimized to maximize image quality. To convert the acquired raw OCT data into a depth profile, an inverse Fourier transform relating the wavenumber k to the physical distance is used. Even with a properly calculated tuning curve 166, the acquired spectral fringe pattern is typically not perfectly linear in k, even when the tuning curve is periodically recalculated. K-linearization uniformly resamples the raw interferometric data in k-space, which improves axial resolution.

[0071] This resampling requires a frequency reference across the sweep period to accurately monitor the instantaneous frequency of the laser for each captured line of the line field sensor 230. In conventional systems, this frequency reference is often obtained using a separate fiber interferometer or etalon that receives a portion of the laser's emission, and the transmission of the interferometer or etalon is monitored using a separate detector and analog-to-digital converter. In other cases, especially when the laser exhibits good scan-scan repeatability, the sample is replaced with an etalon or other spectrally periodic reference, and the k-linearization subsequently applied to subsequent scans is determined.

[0072] Figure 3 A linear field sensor 230 and its pixel P linear array are shown. As previously mentioned, sensor 230 often has hundreds (such as at least 512) pixels, and often at least 1024 or 2048 pixels. In the example shown, the linear array has n pixels in an array of single-pixel widths, but it can definitively be several pixel widths, such as between 2 and 10 pixel widths or more. Pixels can often be binned along this horizontal axis to achieve higher sensitivity.

[0073] In this invention, a frequency reference 310 is placed before one or more pixels P of the line field sensor 230, and is referred to as a reference pixel RP. Preferably, the frequency reference generates a predetermined pattern across the scanning band of the laser 100. A frequency reference that generates a predetermined and periodic pattern is preferred. Here, an etalon (such as a microscope slide or transmission substrate) is used as the periodic frequency reference 310. It is placed over one or more pixels, typically at the beginning or end of the line of the pixel P of the line field sensor. The covered pixels are referred to as reference pixels RP1-RP6 of the linear pixel array 312 of the sensor 230. The remainder of the array 312 is used to detect interference data or fringe patterns from a patient or sample.

[0074] Other frequency references can be used, such as thin-film interference filters, photonic crystals, and etched diffraction gratings, to name a few examples.

[0075] It should be further pointed out that the frequency reference 310 can be placed Figure 1 Other locations in the optical train shown. For example, in another example, the periodic frequency reference 310 is repositioned from in front of the line-scan sensor 230 before or within the reference arm of the beam splitter 210. Yet another example is a wedge-shaped window, placed close to or as part of the reference mirror 226. Generally, the periodic frequency reference 310 is positioned after the laser 100 and preferably after the line-forming optics 208, as part of the collimated beam in the intercept system 200. That portion of the beam is then imaged on one or more defined pixels or reference pixels P1-P6 of the linear pixel array 312 of the sensor 230, often at one of the points in the array.

[0076] Return to reference Figure 2 These reference pixels detect the frequency reference 310, i.e., the periodic transmission 316 of the etalon, when the laser 100 is tuned by its sweep across the scanning band.

[0077] By analyzing the response of the reference pixel RP during the laser sweep across the scanning band, the SOM 235 resolves the nonlinearities in the sweep of laser 100, and these nonlinearities are used to calculate a resampling curve for resampling the interferometric data to be linear in k, as part of a k-linearization preprocessing step. Furthermore, in some examples, the determined nonlinearities in the sweep of laser 100 are used to further update the tuning function stored in the tuning curve module 166.

[0078] The k-linearization process uniformly resamples the original interferometric data in the k-space, which improves axial resolution. In one example, the resampling curve r[j] is specified by providing the coefficients of a third-order polynomial. Resampling curves are often stored as lookup tables in or maintained by the SOM 235. The resampling curve is the original data array S. raw Each index j in [j] is assigned an index j', i.e., j' = r[j]. To obtain the original data array S linearized by k... k [j], the value at index j' needs to be interpolated and remapped to the array position at index j.

[0079] The step of performing k-linearization involves interpolating the interferometric data from each laser sweep using the SOM 235. Several different methods are common, including linear, cubic spline (Catmull-Rom spline), and Lanczos methods. These methods represent a trade-off between speed and accuracy, with linear being the fastest and Lanczos being the most accurate.

[0080] In different embodiments, the resampling curve is determined by the SOM 235 at different frequencies. A new resampling curve can be determined for each sweep of the laser 100, and the new resampling curve is used by the SOM 235 to linearize the interferometric data of that sweep and / or subsequent or previous sweeps of the laser. In other examples, the resampling curve is resolved for every nth sweep of the laser 100, and the linearization implemented in the resampling curve is used for subsequent n sweeps. In another example, the nonlinearity in the sweeps of the laser 100 is determined and compared with the expected lowest acceptable linearization parameter. When the nonlinearity of the sweep has degraded to below the lowest acceptable linearization parameter, a new resampling curve is calculated and then applied to subsequent sweeps of the laser and / or retrospectively applied to previous sweeps. In yet another example, the determined nonlinearity is used to reprogram the laser controller 232 to sweep the laser 100 more linearly.

[0081] Figure 4A , Figure 4B , Figure 4C and Figure 4D This is a flowchart illustrating different methods for calculating and applying new resampling curves or more linear tuning for k-linearization in the OCT system 200. The resampling curves are used by the SOM 235 to correct for nonlinearity in the sweep of the laser 100, thereby ensuring accurate depth profiles in OCT imaging.

[0082] in short, Figure 4A This illustrates a process where, for example, the resampled curve is periodically recalculated as part of a self-calibration process. In contrast, Figure 4BThe process for determining a resampling curve for each sweep of the laser is illustrated. This continuous resampling curve calculation is useful in OCT applications requiring phase stability, such as those used for OCT angiography. In the relevant process, a new resampling curve is calculated after a predetermined number of sweeps (such as 100 or 1000 or more sweeps of the laser). Figure 4C The process of calculating a new resampling curve when the laser's sweep linearity drops below a predetermined linearization threshold is illustrated. Finally, Figure 4D The process is illustrated in which the tuning function of the laser, stored in the tuning curve module 166 of the laser controller, is additionally updated in response to determining that the sweep linearity of the laser has fallen below a predetermined linearization threshold.

[0083] Periodic resampling curve recalculation

[0084] Figure 4A Process 400-1 is depicted, in which the resampling curve is periodically recalculated, for example, as part of a self-calibration process. The process begins at step 410, where n interference data lines are captured as the laser 100 sweeps across its scan band. This interference data, used for the B-scan, is collected by the line field sensor 230.

[0085] In step 420, background is preferably removed. In one example, a rolling average with a user-adjustable window size is used. Subsequently, the estimated background is subtracted from the spectrum. This step effectively eliminates the DC component in the resulting OCT image, which proves particularly beneficial when using an OCT sweep source that produces sweeps with different intensities.

[0086] In step 440, the resampled curve is used to perform k-linearization as described above.

[0087] In step 450, a window function and dispersion correction are applied. In windowing, the original data is multiplied by the window function, which sets the signal outside a predefined interval to zero to reduce sidelobes. Dispersion correction is another common correction where wavenumber-dependent phase shifts are corrected due to the presence of dispersive media of varying lengths in the path.

[0088] Finally, in step 460, a B-scan is created from the sweep by performing an inverse Fourier transform to generate a depth profile across the B-scan.

[0089] These steps may be repeated for a set number of B-scans, a set number of patient scans, or for a set time period.

[0090] After repeating the process according to one or more scan or patient programming parameters or programmed time periods, in step 430, SOM 235 analyzes the reference pixel RP (such as... Figure 3The response of RP1-RP6 in the laser scan is used to determine a new resampling curve. The periodic transmission pattern 316 from the frequency reference 310 is processed to identify nonlinearities in the laser scan, and a new calculated resampling curve is created by the SOM 235 to perform k-linearization on the next captured interferometric data, thereby aligning data points to a uniform k-space grid for the next scan and / or the next patient based on the new resampling curve. The newly calculated resampling curve can also be retrospectively applied to improve previous B-scans.

[0091] Therefore, this process is repeated periodically as part of the system's self-calibration protocol.

[0092] Continuous resampling curve calculation

[0093] Figure 4B The diagram illustrates process 400-2, in which the resampling curve is recalculated for each sweep of laser 100. This continuous process is particularly suitable for phase-stable applications, such as OCT angiography.

[0094] In step 405, as laser 100 is tuned across its scan band, sensor 230 captures interference data lines. In step 430, the response of reference pixel RP is analyzed to determine the resampling curve required for that particular sweep.

[0095] In step 440, the interferometric data is k-linearized using the determined resampling curve to ensure accurate alignment in k-space.

[0096] Repeat these steps until the scanning band across the laser has been swept to collect the required interference data for the entire B-scan.

[0097] The process then proceeds to step 450, where a window function is applied and dispersion is corrected, and step 460, where an IFFT is performed to generate a B-scan. This method ensures real-time calibration and imaging accuracy for each scan.

[0098] Threshold-based recalculation

[0099] Figure 4C Method 400-3 is described, in which a new resampling curve is calculated only when the linearity of the laser sweep drops below a predetermined threshold. The process begins at step 410, capturing n interference data lines for the B-scan.

[0100] The background is removed in step 420.

[0101] In step 430, SOM 235 analyzes the response of the reference pixel RP to assess the scanning linearity of the laser and the correctness of the current resampling curve. In step 432, SOM 235 evaluates whether the existing current resampling curve is acceptable based on a predefined linearization threshold.

[0102] If the existing linearization is acceptable (yes), the process jumps to step 440, where k-linearization is performed using the existing / previous resampling curve. If the linearization is unacceptable (no), then a new resampling curve is computed in step 430 and instantiated as the current resampling curve. In step 440, the current resampling curve is applied to perform k-linearization.

[0103] After k-linearization, a window function and dispersion correction are applied in step 450, and an IFFT is performed in step 460 to generate a B-scan. This threshold-based method recalibrates the system only when necessary, thus balancing computational efficiency with image quality.

[0104] Tuning function update

[0105] Figure 4D The threshold-based recalibration process is extended by including updating the tuning function of the laser stored in the tuning curve module 166. The process begins at step 410, capturing n interference data lines during laser sweep.

[0106] The background is removed in step 420.

[0107] In step 430, SOM 235 analyzes the response of the reference pixel RP to determine the required resampling curve and thus the tuning linearity of the laser. Step 432 evaluates whether the existing linearization is acceptable. If it is acceptable (yes), then the process continues with k-linearization in step 440.

[0108] If linearization is unacceptable (No), then the tuning function is updated in step 438. This involves recalculating the tuning parameters based on the laser's tuning transfer function and storing them in the tuning curve module 166 of the laser controller 232. This update ensures improved laser sweep linearity for future scans.

[0109] Once the resampling curve is applied, a window function is implemented in step 450, and an IFFT is performed in step 460 to create a B-scan. This method dynamically adjusts the laser's tuning parameters to maintain system performance over time.

[0110] As in Figure 4A – Figure 4DAs shown, these processes provide a general framework for managing k-linearization in the OCT system 200. The ability to calculate and apply resampling curves in real time or as needed ensures high-quality depth profile and tomographic images, thus adapting to diverse operational requirements and laser characteristics.

[0111] While the invention has been specifically shown and described with reference to preferred embodiments thereof, those skilled in the art will understand that various changes in form and detail may be made therein without departing from the scope of the invention as covered by the appended claims.

Claims

1. An optical coherence tomography (OCT) system, the system comprising: A swept-frequency laser source, the swept-frequency laser source being configured to emit light having a wavelength tuned across the scan band; A line field sensor comprising a linear array of pixels configured to detect interference signals corresponding to light scattered from a sample and a reference light; A frequency reference is deployed to interact with a portion of the light emitted by the swept laser source and generate a reference pattern across the scan band, the periodic reference pattern being detected by pixels in the line field sensor; Computer, the computer is configured to: (a) Determine a resampling curve based on the reference pattern to correct the nonlinearity in the tuning of the swept laser source; (b) Apply a resampling curve to linearize the interference signal detected by the line field sensor with k; as well as (c) Perform an inverse Fourier transform on the k-linearized interferometric signal to generate a depth profile; as well as A display configured to render a tomographic image generated from a depth profile.

2. The system of claim 1, wherein the frequency reference includes a etalon configured to generate a periodic interference pattern when the swept laser source is tuned across the scan band.

3. The system of any one of claims 1 and 2, wherein the etalon is positioned such that it covers only a portion of a linear array of pixels, thereby defining a set of reference pixels dedicated to detecting the periodic reference pattern.

4. The system according to any one of claims 1-3, wherein the reference pixel is located at one end of a linear array of pixels.

5. The system of any one of claims 1-4, wherein the swept laser source comprises a cat's-eye configuration, the cat's-eye configuration comprising a bandpass filter and an angle control actuator for tilting the tuned filter to achieve wavelength sweeping.

6. The system of any one of claims 1-5, wherein the computer is configured to periodically recalculate the resampling curve after a predetermined number of sweeps, thereby maintaining or improving the k-linearity over time.

7. The system of any one of claims 1-6, wherein the computer is further configured to update the tuning function of the swept laser source.

8. The system of any one of claims 1-7, further comprising a scanning mechanism configured to move the illumination line across the sample, wherein the tomographic images are formed from a series of k-linearized depth profiles acquired at different lateral positions.

9. An optical coherence tomography system, the system comprising: A frequency-sweeping laser source, the frequency-sweeping laser source including an intracavity mechanical tuning mechanism and a tuning curve module that stores predefined tuning functions to control the tuning of the frequency-sweeping laser source; A line field sensor, configured to detect an interference signal generated by a frequency reference and a reference signal during the tuning of the frequency-sweeping laser source; The controller is configured to: (a) Evaluate the linearity of the tuning based on the reference signal; (b) Determine the resampling curve to compensate for nonlinearity in wavelength tuning; as well as (c) Update the predefined tuning functions in the tuning curve module; as well as A processor configured to apply the resampling curve to k-linearize the interferometric signal and generate a depth profile for imaging.

10. The system of claim 9, wherein the frequency reference includes a etalon positioned to cover only a portion of a linear array of pixels of the line field sensor, such that at least one pixel is designated for detecting a periodic reference pattern indicating tuning linearity.

11. The system of any one of claims 9 and 10, wherein the controller is configured to update the predefined tuning function after determining that the linearity of the tuning has dropped below a predefined threshold, thereby ensuring improved wavelength sweep consistency in subsequent scans.

12. A method for optical coherence tomography imaging, comprising: The interference signal corresponding to the light scattered from the sample and the reference light was captured during the tuning of the frequency-sweeping laser source using a line field sensor. A subset of pixels in the line field sensor is used to detect a periodic reference pattern generated by the interaction of a frequency reference with light from the swept laser source; A resampling curve is determined based on the periodic reference pattern to correct the nonlinearity in the tuning of the swept laser source; The resampling curve is applied to linearize the interference signal with k. as well as Perform an inverse Fourier transform on the k-linearized interferometric signal to generate a depth profile for computed tomography imaging.

13. The method of claim 12, further comprising updating a predefined tuning function for the swept laser source based on the periodic reference pattern, thereby improving the linearity of subsequent wavelength sweeps.

14. The method of any one of claims 12 and 13, wherein detecting the periodic reference pattern comprises using an etalon that covers only one or more selected pixels of the line field sensor, enabling the simultaneous acquisition of the interference signal and the reference signal with a single detector.

15. A method for calibrating a swept-frequency laser source in an optical coherence tomography system, comprising: The reference signal generated by the frequency reference is captured during wavelength tuning of the swept laser source; The reference signal is analyzed to evaluate the linearity of wavelength tuning; Determine the resampling curve to compensate for nonlinearity in wavelength tuning and / or update the tuning function stored in the swept laser source; The resampling curve is applied to linearize the interference signal captured by the OCT system with k. as well as Tomographic images are generated based on the k-linearized interferometric signal.

Images