A system and method for improving optical coherence tomography (OCT) image resolution using K-linearization (KL) and variance correction (DC).

Abstract

Aspects of the present disclosure relate to systems, methods, and algorithms for performing wavenumber linearization and dispersion correction in optical systems that do not require hardware modifications, empirical tuning, precise mirror alignment, and can be performed in real time at low computational cost. A one-time calibration process can generate spectra or calibration standards, including wavenumber linearization standards, dispersion correction, and spectrally flattened spectra, that can be used to correct optical coherence tomograms in real time.

Description

[Technical Field]

[0001] This disclosure generally relates to the field of vascular imaging and data acquisition systems and methods. In particular, this disclosure relates to methods for improving the resolution of images acquired in optical coherence tomography systems.

[0002] [Cross-reference of related applications] This application asserts the benefit as of the filing date of U.S. Provisional Patent Application No. 63 / 146,904, filed on 8 February 2021, the disclosures of which constitute part of this specification by reference. [Background technology]

[0003] Optical coherence tomography (OCT) is an imaging technique that uses light to capture cross-sectional images of tissue at the micron scale. OCT can be used both in situ and in systems located outside the organism being sampled.

[0004] The axial resolution of OCT images can decrease, more specifically in optical coherence tomography systems in the Fourier domain, along with chromatic dispersion and the spread of sidelobe artifacts due to nonlinear wavenumber sampling. Many OCT systems sample interference patterns at non-uniform wavenumber (k) intervals. This can introduce "chirp" or noise into the signal, which may depend on the path length differences of the OCT signal.

[0005] Controlling both nonlinear wavenumber sampling (also known as k-linearization) and chromatic dispersion through wavenumber linearization is a crucial step in maintaining the resolution of OCT images.

[0006] Chromatic dispersion is unique to optical systems and can be controlled by carefully controlling the optical fibers and other components within the system. Wavenumber linearization presents a more challenging technical issue and is performed by applying numerical methods, such as numerical rescaling based on fringe signals from mirrors within the OCT system, or by using an external sampling clock (k-clock) as used in the current OPTIS integrated system commercially available from Abbott Vascular. While an external k-clock can attempt to linearize the nonlinear sampling interval of the wavenumber, the accuracy of wavenumber linearization is limited in the system due to improper adjustment of k-clock delay and chromatic dispersion. Nonlinear sampling due to k-delay errors causes imaging artifacts such as reduced axial resolution and enlarged sidelobes. Even with precise adjustment of the k-delay, resolution can be directly reduced by slight system dispersion. In systems without dispersion, the k-clock delay can be precisely adjusted by placing mirrors on the sample arm and minimizing the linewidth. However, in systems with dispersion, the combination of k-linearization and chromatic dispersion results in k-delay adjustments that only achieve optimal resolution at specific depths. Furthermore, even when a swept-source OCT engine is optically optimized, electronic components such as bandpass filters and cables can introduce unexpected wavenumber delays to the external sampling clock.

[0007] An algorithm has been proposed to digitally compensate for inappropriate k - delays by calculating the "true" k - linearized spectrum that needs to be separated from system dispersion. Then, the fringe data is resampled by interpolating the "true" k - spectrum to linearize the sampling in k - space. The system dispersion is then compensated by multiplying the interference signal by a counter - dispersive phase before the fast Fourier transform (FFT). Generally, calculating the system dispersion phase is an iterative process of optimizing the second - and third - order dispersion coefficients based on the sharpness of the OCT mirror signal. However, the dispersion order varies with the OCT system. Since prior knowledge of the dispersion order is required, the iterative process in commercial OCT systems may have limited feasibility.

[0008] Algorithms have been proposed to generate a “true” remapping table (wavenumber spectrum) to compensate for k - delay. However, such algorithms have hardware requirements that need measurements of system dispersion. Therefore, especially in commercial products, implementing the remapping table has become difficult or impractical. Other methods proposed to compensate for k - delay bias and system dispersion are computationally intensive, require processing times of tens of minutes, and need accurate alignment of the imaged mirror positions, making them impossible to implement. Recently, an algorithm has been proposed that generates a remapping table in a system with dispersion and then derives and compensates for the system dispersion. This method is simple and effective and does not require changes to the hardware setup. However, since the reported useful imaging depth after correction is only 1 mm (in air), if the accuracy of the calculated remapping table is insufficient, it cannot be used for actual cardiovascular applications. To achieve optimal imaging performance across the entire imaging depth, one of the main focuses of the present disclosure is to propose a polynomial optimization process to improve the accuracy of the “true” remapping table and show an improvement in SNR and resolution at deeper depths. To date, the algorithms described above have been effective in compensating for k - non - linearity and system dispersion in a controlled research environment but have various issues that prevent their realization in commercial OCT systems.

Summary of the Invention

[0009] In part, the present disclosure relates to systems and methods that simultaneously correct non - linear sampling and chromatic dispersion problems without the need for additional hardware modifications, prior knowledge of the dispersion order, accurate mirror alignment, and computational costs associated with other systems and methods proposed heretofore.

[0010] In part, this disclosure relates to a system and method for acquiring calibration data of an OCT system at various calibration depths, and to a calibration process for optimizing spectra using wavenumber linearization or k-linearization (KL), dispersion correction (DC), and spectral flattening (SF).

[0011] The disclosed aspects of the technology include algorithms for compensating for k-delay bias and system variance without prior knowledge of additional hardware and system variance. Furthermore, the disclosed aspects of the technology do not require precise alignment of mirror positions by using the proposed phase-shift process.

[0012] Aspects of the disclosed technology enable an increase in the useful imaging depth of spectral domain OCT systems and wavelength-swept OCT systems compared to the 14% and 20% limits of currently published methods. Aspects of the disclosed technology include a polynomial optimization process by using three or more mirror sampling positions at different depths, having at least two depths within the positive imaging plane, to optimize the k-linearized spectrum and maintain improved resolution and SNR over the entire depth of more than 80% of the Nyquist depth. As one example, the disclosed technology enables imaging depths beyond 6 mm with high accuracy, improving OCT image resolution and SNR in applications such as detecting pericoronary calcium plaques (which require a maximum depth of 6 mm).

[0013] The disclosed aspects of the technology can be implemented in the order of seconds without requiring oversampling or any upscaling, enabling real-time product use. Furthermore, due to its high computational efficiency, the proposed algorithm can be implemented on live samples imaged using OCT technology.

[0014] The disclosed aspects of the technology eliminate the need for an external k-clock to compensate for nonlinear sampling in k-space. As one example, the proposed method improves the system signal-to-noise (SNR) ratio by 3 to 5 decibels (dB) by using a 1 GHz internal sampling rate of an analog-to-digital (A / D) card that is faster than the k-clock frequency. As the internal sampling rate of the analog-to-digital card increases, system noise is averaged incoherently, but signals are added coherently, thus increasing the number of sampling points that can improve the image contrast after the Fourier transform.

[0015] The disclosed aspects of the technology enable use in existing commercially available intravascular OCT products through software or method modifications, improving axial resolution by compensating for k-delay bias, system variance and spectral flattening, and providing extended high-resolution depth and low computational cost by using optimized KLDC spectra.

[0016] The disclosed aspects of the technology enable the use of internal sampling provided by an analog-to-digital (A / D) card, which allows for the elimination of the k-clock and improves system SNR and imaging depth. This is particularly ideal for future high-speed OCT systems, as it requires faster sampling rates to reach signals from greater depths.

[0017] Aspects of the disclosed technology include a method for processing optical signals to improve the resolution of optically acquired images. The method may include calibrating an optical system, which includes (i) acquiring at least a first and a second mirror measurement by one or more processors, the first and second mirror measurements being collected from both sides of a zero-delay line, and each mirror measurement being an interferogram or a signal or system impulse response; (ii) acquiring the amplitude and phase of the first and second mirror measurements by one or more processors; (iii) resampling at least one of the first and second mirror measurements to be in (a) linear wavenumber (k) space or (b) linear wavelength to generate a resampled mirror measurement; and (iv) calculating an optimized fit of the resampled mirror measurement using a function. In some examples, two mirror measurements are used, but in other examples, additional mirror measurements may be used. For example, calibration can be based on 3, 4, tens, hundreds, or any number of Miller measurements.

[0018] The aspects of the disclosed technology may include any combination of the following, including a method: The method may include calibrating an optical system, the calibration including (i) acquiring at least a first mirror measurement and a second mirror measurement by one or more processors, the first and second mirror measurements being collected from both sides of a zero delay line, and each mirror measurement being an interferogram or a signal or system impulse response; (ii) acquiring the amplitude and phase of the first and second mirror measurements by one or more processors; (iii) resampling at least one of the first and second mirror measurements to be in (a) linear wavenumber (k) space or (b) linear wavelength to generate a resampled mirror measurement; and (iv) calculating an optimized fitting of the resampled mirror measurement using a function. The function used to resample the data may be one of the following: a polynomial function, a cubic or other spline fitting function, a radial basis function, or a piecewise function. K-linearization can be performed on the raw interferogram of Miller measurements with or without k clocks to produce a resampled interferogram, where the resampled Miller measurements are in k-space. Optimization of the K-linearization function can be based on a polynomial fitting order of order 1 to n, where n is a fixed integer. The polynomial fitting order that minimizes the sum of (i) the area of ​​the point intensity distribution function and / or (ii) the full-width half max (FWHM) resolution for each Miller measurement can be determined as part of the method. The optimized polynomial can be determined based on two or more Miller measurements.The dispersion spectrum or criterion can be calculated for at least one mirror measurement, and compensation can be calculated. Dispersion compensation, dispersion coefficient, or dispersion criterion can be calculated using two mirror positions, one on each side of the "zero delay line". Spectral flattening can be performed for at least one mirror measurement by using the envelope calculated from a specific mirror measurement. The spectral envelope calculated during spectral flattening can be saved. Each mirror measurement in a set of multiple mirror measurements may be a system impulse response. Calibration criteria for (i) k-linearization (KL), (ii) dispersion correction (DC), and (iii) spectral flattening (SF) can be saved. An OCT signal or interference signal can be received from a specimen. Using the KL calibration criterion, a linear k-interpolation can be generated to k-linearize the received OCT signal or interference signal, after which a new fringe may be generated or the interferogram may be resampled. Dispersion correction can be performed on the k-linearized OCT signal or interference signal using the DC calibration criterion. The envelope of the k-linearized and variance-corrected OCT signal or interference signal can be removed. The OCT signal or interference signal can be converted into an OCT image for display on a display.

[0019] The aspects of the disclosed technology include methods. The methods may include generating an OCT image from an OCT signal, and the methods include acquiring an OCT signal corresponding to a specimen; linearizing the OCT signal by wavelength (k) to generate a k-linearized (KL) OCT signal; dispersion-correcting (DC) the KL OCT signal to generate a KL-DC OCT signal; spectrally flattening (sf) the KL-DC OCT signal to generate a final KL-DCsf OCT signal; and performing post-processing on the final OCT signal to generate an OCT image. The KL OCT signal, KL-DC OCT signal, and final KL-DCsf OCT signal are based on a calibration criterion generated in a calibration step. The OCT signal may be an interferogram. The calibration criterion may be generated in a calibration step using at least a first and second mirror measurement, the first and second mirror measurements being collected from both sides of a zero-delay line, and each mirror measurement being an interferogram or signal or system impulse response.

[0020] Embodiments of the disclosed technology include a system for displaying a set of images. The system may include a memory for storing image data and time-varying data corresponding to a subject, and one or more processors for communicating with the memory. One or more processors may be operable to acquire first and second mirror measurements, the first and second mirror measurements being collected from both sides of a zero-delay line, and each mirror measurement being an interferogram or signal or system impulse response; to calculate the amplitude and phase of the first and second mirror measurements; to resample at least one of the first and second mirror measurements to (i) linear wavenumber (k) space and (ii) linear wavelength to generate a resampled mirror measurement; and to calculate an optimized fitting of the resampled mirror measurement using a function. The system may be configured to perform any of the steps of the above-described method.

[0021] Embodiments of the disclosed technology include a system for performing optical coherence tomography on a sample. The system may comprise a light source, a reference mirror with a variable position, a display, a memory for storing image data corresponding to the sample, and one or more processors communicating with the memory. The one or more processors may be operable to acquire at least one calibration parameter each time the system is started, acquire a signal from the sample, apply at least one calibration parameter to the acquired signal, wherein the calibration parameter is at least one of wavelength linearization, dispersion correction, or spectral flattening parameters, and generate an image from the corrected acquired signal. The system may further comprise an optical switch and a calibration mirror configured to generate a calibration signal. The system may be configured to perform any of the steps of the method described above.

[0022] It should be understood that the different aspects and embodiments disclosed herein can be combined together, either as a whole or in part, as necessary. Thus, each example disclosed herein can be incorporated into each aspect to varying degrees as necessary for a given implementation. Furthermore, various software-based tools for addressing medical imaging problems and other related issues and problems, and some of those mentioned above, can be used, without limitation, for medical applications and other applications for displaying information related to stents and blood vessels and their two-dimensional and three-dimensional views. Other features and advantages of the disclosed examples will become apparent from the following description and accompanying drawings.

[0023] This disclosure relates to different embodiments, examples, and other features cited herein, but it is understood that each of the above disclosed herein can be combined together, in whole or in part, as necessary. Thus, each example disclosed herein can be incorporated into each embodiment to varying degrees as necessary for a given implementation. Furthermore, the various methods and techniques described herein can be used with various imaging modalities.

[0024] Other features and advantages of the disclosed example will become apparent from the following description and accompanying drawings. [Brief explanation of the drawing]

[0025] [Figure 1A] This is a schematic diagram of an imaging and data acquisition system according to the aspects of this disclosure. [Figure 1B] This figure shows an exemplary positioning of a reference mirror according to an embodiment of the present disclosure. [Figure 1C] This is a schematic diagram of an imaging and data acquisition system according to the aspects of this disclosure. [Figure 1D] This is a block diagram of an exemplary calibration mirror according to an aspect of the present disclosure. [Figure 2] This figure shows an example computing device that performs imaging and data acquisition according to the embodiments of this disclosure. [Figure 3] This is a flowchart of the method according to the aspects of this disclosure. [Figure 4A] This figure shows the characteristics of an interferogram obtained according to the embodiments of this disclosure. [Figure 4B] This figure shows the mode of nonlinear effects due to system dispersion, based on the curves of the k spectrum calibrated by the KL process and the ideal linearized k spectrum. Figure 4B also shows the effective correction for the uncalibrated k spectrum by applying the KL process. The dashed line represents the optical phase calculated after the k linearization process. [Figure 4C]This figure shows the intensity characteristics at various imaging depths, depending on whether or not k-spectral optimization is performed, according to the embodiments of this disclosure. [Figure 4D] This figure shows the raw dispersion phase and the polynomial fit of the dispersion phase according to the embodiments of this disclosure. [Figure 4E] This figure shows the KLDC-corrected fringe and the envelope calculated for the corrected fringe according to an aspect of this disclosure. [Figure 5] This is a flowchart of the method according to the aspects of this disclosure. [Figure 6A] This figure shows the raw optical signal and the k-linearized optical signal according to an aspect of this disclosure. [Figure 6B] This is an interferogram and a magnified view of the corresponding intensity resulting from mirror reflection within a sample arm, according to an aspect of this disclosure. [Figure 6C] This figure shows the raw fringe and the before and after spectral flattening according to an aspect of this disclosure. [Figure 6D] This figure shows various aspects of the full width at half maximum (FWHM) plotted for various depths for k-clock data, KL data, KLDC data, and KLDCsf data according to the embodiments of this disclosure. [Figure 7] This figure shows a method for acquiring intensity images according to an aspect of this disclosure. [Figure 8] This figure shows an example of improvement in the width and resolution chart of the point spread function (PSF) according to the embodiments of this disclosure. [Figure 9] This figure shows the characteristics of PSF with strong dispersion at various imaging depths, depending on the presence or absence of KLDCsf technology, according to the aspects of this disclosure. [Figure 10] This figure shows the k-delay adjustment and the FWHM and signal-to-noise ratio without strong dispersion, using k-clock data, KLDCsf data, and KLDCsf data having an optimized k spectrum, according to an aspect of the present disclosure. [Figure 11]This figure shows the intensity of the a-line using KLDCsf with and without an optimized k-spectrum, according to an aspect of this disclosure. The "A-line" or "axial line" corresponds to a single wavelength sweep of the laser, which corresponds to a one-dimensional line scan. [Figure 12] This figure shows the data sampled linearly over time (wavelength) without using a k-clock, and the data processed using the KLDCsf method, according to the embodiments of this disclosure. [Figure 13] This disclosure describes an embodiment of FWHM that uses data resampled using k-clocks at various imaging depths and uses A / D card internal sampling using the KLDCsf technique. [Modes for carrying out the invention]

[0026] In part, this disclosure relates to a calibration process capable of calculating k-linearized (KL), variance, and spectral flattening spectra. The calibration process can provide a set of calibration spectra. An OCT signal, including raw fringe data with a mirror in the sample arm, can be acquired separately from the positive and negative planes by adjusting the reference arm path distance. Calibration spectra can be calculated based on the phases extracted by using a Hilbert transform process.

[0027] In part, this disclosure relates to utilizing acquired calibration spectra to modify or correct an interferogram before fast Fourier transform (FFT) processing. Using the calibration spectra, future fringe data acquired by the system in the wavenumber domain can be linearized for an improved interferogram. In some examples, wavenumber sampling can be linearized by interpolation using the calibration spectra (k-linearization). Variance can be corrected by applying a Hilbert transform and multiplying it by the variance spectrum of the k-linearized interferogram.

[0028] In part, this disclosure may be further related to suppressing sidelobe artifacts caused by non-uniform laser intensity during spectral sweeping by flattening the spectral envelope of the laser source to optimize the laser bandwidth.

[0029] In part, aspects of the present disclosure provide systems and methods for separately correcting not only k-spectral nonlinearity but also dispersion and asymmetric laser sweep intensity by using a calibrated spectrum with a signal processing step. The disclosure enables optimized optical resolution to be maintained across the entire imaging depth. Furthermore, the algorithms, methods, and systems described herein can also function with or without a k-clock, providing an option to improve the system SNR by using an internal sampling rate of the digitizer that is typically faster than the maximum frequency of the k-clock.

[0030] In some aspects of this disclosure, the OCT imaging engine includes an internal reference reflector or calibration mirror that automatically receives information corresponding to the quality of the OCT imaging engine. The reference reflector may be located inside the sample arm of the OCT imaging engine and may be used to receive optical signals for performing various diagnostic processes that quantify the performance of the OCT imaging engine. For example, the OCT imaging engine can self-assess its performance by calculating system performance data which may include one or more of the following: point intensity distribution function (PSF), full width at half maximum (FWHM), noise level, signal-to-noise ratio, and system dynamic range. The reference reflector may be coupled to the rest of the OCT imaging engine through an optical switch, allowing the OCT imaging engine to switch between imaging mode and self-inspection mode. The OCT imaging engine can use the internal reference reflector to sample optical signals to generate a calibration spectrum. Because the reference reflector is internal, the OCT imaging engine can perform self-inspection and calibration automatically without requiring user input or an external device to be initially connected to the system.

[0031] As used in this disclosure and as understood by those skilled in the art, a system impulse response or impulse response is a response to a short input signal that is an impulse. In an optical system, the system impulse response may be a response obtained from an impulse of light on a sample. In some examples, the system impulse response may be a reflection obtained from a single mirror in a sample arm (as described below). In some examples, the system impulse response may be obtained from a wavelength-swept laser.

[0032] As used in this disclosure, an interferogram or interference pattern may be a pattern formed by wave interference, such as by the interference of light waves from the reference arm and sample arm of an OCT system. An interferogram can also be considered a time-varying signal that can be converted using an analog-to-digital converter.

[0033] (Example system) OCT is a catheter-based imaging modality that uses light to peer into the coronary artery wall and generate images for research. Utilizing coherent light, interferometry, and micro-optics, OCT can provide video-rate in vivo tomography of affected vessels with micrometer-level resolution. By observing subsurface structures with high resolution using fiber optic probes, OCT is particularly useful for minimally invasive imaging of internal tissues and organs. This level of detail made possible with OCT allows users to diagnose and monitor the progression of coronary artery disease.

[0034] OCT imaging of parts of a patient's body provides a useful diagnostic tool for physicians and others. For example, intravascular OCT imaging of coronary arteries can reveal the location of narrowing or stenosis. This information helps cardiologists choose between invasive coronary artery bypass surgery and less invasive catheter-based procedures such as angioplasty or stent delivery. Although a popular option, stent delivery carries its own associated risks.

[0035] Figure 1A shows an exemplary OCT system 100 according to an aspect of the present disclosure. In some examples, the OCT system 100 is an optical interferometer or incorporates an optical interferometer. Those skilled in the art will understand that while one configuration of the OCT system is shown in Figure 1, variations of the system and different embodiments, such as in vivo OCT systems, are within the scope of the present disclosure. The direction of the light or electrical signal is indicated by the arrows in Figure 1.

[0036] The light source 110 may be a low-coherence light source capable of acquiring resolution at the sub-micrometer level. The light source can generate light within the visible wavelength range and light beyond that wavelength range. In some examples, an ultra-wide wavelength output of light is desirable. In some examples, a laser can be used as the light source. In yet another example, a light-emitting diode can be used as the light source. In some examples, the light generated by the light source 110 can be transmitted through a collimation lens.

[0037] Light from the light source 110 can be sent to a beam splitter 120. The beam splitter 120 may be an optical device that splits the light beam from the light source 110 into two or more beams. The light split by the beam splitter can be directed to a reference mirror 130 and a sample 140. An embodiment of the reference mirror 130 will be further described with reference to Figure 1B. The sample 140 may be an organic or other sample on which OCT can be performed. In some examples, the sample 140 may be examined internally, as in the case of an in vivo OCT scan. Light may be reflected from both the reference mirror 130 and the sample 140 through an optical path that crosses the beam splitter and directs the light to a photodiode 150. The photodiode 150 may be a semiconductor or other device that converts light into an electric current, enabling the detection of light. When a photon or wave of light is incident on the photodiode, an electric current or another electrical signal is generated within the photodiode. The photodiode 150 may include multiple optical filters, lenses, or other components to focus light and increase the signal-to-noise ratio. The signal generated in the photodiode 150 can be converted from analog to digital and processed by the digital signal processor 160. The digital signal processor may be a dedicated microprocessor or integrated chip having an architecture and / or software optimized for the operational needs of digital signal processing. In some examples, the digital signal processor 160 may enable the information generated in the photodiode 150 to be processed and sampled by the k clock signal in 191 and converted into an image to be displayed on the display 170. As further described below, the digital signal processor can perform one or more of the steps described below by optimizing the image generated from studying or observing the sample 140. The display 170 can display the image related to the sample 140. In non-limiting examples, the display 170 may be a monitor, an OLED screen, an LCD screen, a television, an electroluminescent display, or a quantum dot display.Other specialized screens or displays that facilitate the display of specialized contrast ratios or OCT information can also be used as the display 170.

[0038] The k-clock 190 is also shown in Figure 1A. If the arrival time of the signal from the k-clock output to the A / D card is not synchronized with the signal from the main interferometer to the A / D card, the interferogram cannot be accurately sampled. Synchronization problems can be caused by a mismatch between the length of the optical fiber and the electronic connections after photodiode 191 and photodiode 150, such as the length of a bandpass filter or electrical cable. In some examples, only a few percent, e.g., 2% to 3%, of the light may be transmitted to the k-clock. The k-clock 190 can consist of a 90-degree phase shifter, a zero-crossing detection unit, an XOR gate or OR gate, or any combination of similar elements.

[0039] The beam splitter 180 can also split the light to the k-clock 190 and the beam splitter 120. In some examples, a small amount of light can be transmitted to the k-clock through the beam splitter. The photodiode 191 is similar to the photodiode 150 and is connected to the k-clock, allowing for the analysis of light incident on the photodiode 191. The photodiode 191 can be connected to the digital signal processor 160. Since some of the light passes through the k-clock, that light can be analyzed separately from the light incident on or acquired from the sample 140.

[0040] Those skilled in the art will understand that while various optical components are referenced in Figure 1, equivalent components can be used or substituted for the system 100 described above. In some examples, related optical components such as optical fibers and optical couplers can be used in place of or in conjunction with the components described herein. For example, those skilled in the art will understand that the same or equivalent setup as described with respect to Figure 1A can be achieved by using optical couplers instead of collimation lenses and beam splitters. The light can take a path determined by the optical fiber wire. Using optical fibers and optical couplers can provide a more robust and simpler optical setup used in commercially available OCT applications. As one example, beam splitter 180 and beam splitter 120 may consist of optical fiber couplers that split one input fiber into two output fibers and split the input light into two paths.

[0041] In some examples, various components may be linked, controlled, and communicated through a suitable computing system, such as computing system 200, which will be further described below with reference to Figure 2.

[0042] Figure 1B shows an embodiment of the reference mirror 130. The reference mirror 130 may be a mirror or other reflective surface having high reflectivity and optical properties that enable the reflection of photons. The reference mirror 130 may be included in the reference arm of the OCT system 100. By scanning the mirror in the reference arm with light from the sample 140, an interference pattern can be generated, and an OCT image can be reconstructed from this interference pattern. As shown in Figure 1B, once the mirror is positioned on the sample 140, the reference mirror 130 can be moved to various positions on both sides of the "zero delay line". Various positions such as positions P1, P2, and P3 can move the image of the mirror to different depth pixels. Positions -P1 and P1-P3 may or may not be equidistant with respect to the zero delay line. For simplicity, only -P1 and P1-P3 are shown, but any number of ordered finite positions may exist. For example, there may be an additional position P4 at a positive pixel depth greater than position P3. In one example, the pixel depth may range from +1024 pixels to -1024 pixels. The pixel depth may depend on the k-clock total sample or internal sampling rate of the A / D card, and half of the total number of pixels used before the FFT, according to the zero-padding data length before the FFT. The imaging depth from pixel depth 0 to pixel depth 1024 is {(center wavelength) 2 It is determined by the formula: / (2*laser bandwidth)*(0.5*total number of sample points from the OCT fringe).

[0043] In some examples, the position of the reference mirror 130 can alter the interference pattern generated from the light returning from the sample 140. In other examples, the position of the reference mirror 130 is used to generate information related to the performance of the optical system and may be used in the manner described herein. In some examples, as further illustrated with reference to Figure 6C, the information generated from positions P1 and P2 can be used to calculate spectral wavelength linearized and spectral flattened spectra.

[0044] Figure 1C can show an OCT system (system 199) having a mirror integrated into the sample arm. Figure 1C shows that system 199 may be similar to system 100 but may further include an additional calibration mirror 193 and an optical switch 192. The optical switch 192 can be optically coupled at one end to a beam splitter 120 and at the other end to the sample 140 and the calibration mirror 193. The signal from the calibration mirror 193, together with a variable position reference mirror 130, enables the system to perform the KLDCsf method in an automated or semi-automated process. In some examples, a “self-diagnosis” procedure may be performed each time the system is started or each time a new optical joint is included in or changed in any part of the system.

[0045] Figure 1D is a block diagram of the calibration mirror 193 of Figure 1C, in several examples. The calibration mirror 193 may include an optical connector 205C for connecting the calibration mirror 193 to the switch 192. Alternatively, the calibration mirror 193 may be fusion spliced ​​for connection to the switch 192. The calibration mirror 193 may include an attenuator 210C. The attenuator 210C may have a predetermined attenuation, for example, it may attenuate the reflected signal so that the magnitude of the reflection is similar to the reflection from the sample 140. The calibration mirror 193 may include a flat surface 215C that provides a single-point reflection. The length of the optical fiber may be the same as the optical path length of the sample 140 so that the light reflected from 215C interferes with the reference arm. The single-point reflection may be referred to as the point image intensity distribution function (PSF) and can be processed by the DSP 160 or other components of the OCT imaging engine to obtain information to quantify the OCT imaging engine.

[0046] In some examples, the OCT imaging engine receives mirror measurements using a reference mirror and a calibration mirror. These mirror measurements may include a time-varying amplitude of the interferogram. The system can then use these received measurements to extract the phase of the optical signal as a function of the sampling index.

[0047] Figure 2 shows an exemplary computing system 200. The computing system 200 may include hardware, software, and other modules as further described herein. The following description is intended to provide an overview of device hardware and other operating components suitable for carrying out the methods of this disclosure as described herein, which may be part of the computing system 200. This description is not intended to limit the applicable environments or the scope of this disclosure. Similarly, the hardware and other operating components may be suitable as part of the apparatus described above. This disclosure can be implemented in other system configurations, including personal computers, multiprocessor systems, microprocessor-based electronic devices or programmable electronic devices, network PCs, minicomputers, mainframe computers, etc. This disclosure can also be implemented in a distributed computing environment where tasks are performed by remote processing devices linked through a communication network, such as different rooms in an OCT laboratory or catheterization laboratory.

[0048] Some parts of the detailed description are presented by algorithms and symbolic representations of operations on data bits in computer memory. These algorithmic descriptions and representations can be used by those skilled in the computer and software-related fields. In one example, an algorithm is generally considered herein to be a self-consistent sequence of operations that produce a desired result. Operations performed as method steps or otherwise described herein are operations that require the physical manipulation of physical quantities. These quantities usually, though not necessarily, take the form of electrical or magnetic signals that can be stored, transferred, combined, transformed, compared, and otherwise manipulated.

[0049] Unless otherwise specified, as will be apparent from the following discussion, throughout this explanation, discussions using terms such as “processing,” “computing,” “overlaying,” “searching,” “detecting,” “measuring,” “calculating,” “comparing,” “generating,” “determining,” or “displaying,” or other sets related to Boolean logic or operations, refer to the actions and processes of a computer system or electronic device, which are understood to manipulate data represented as physical (electronic) quantities in the registers and memory of the computer system or electronic device, and to convert it into other data similarly represented as physical quantities in electronic memory or registers or other such information storage devices, transmission devices, or display devices.

[0050] This disclosure also relates, in some examples, to apparatus for carrying out the operations described herein. Such apparatus may be constructed specifically for a required purpose, or it may include a general-purpose computer that is selectively started or reconfigured by a computer program stored in the computer.

[0051] The exemplary systems of this disclosure can be embodied in many different forms, including, but not limited to, any other means including computer program logic for use with a processor (e.g., a microprocessor, microcontroller, digital signal processor, or general-purpose computer), programmable logic for use with a programmable logic device (e.g., a field-programmable gate array (FPGA) or other programmable logic device), discrete components, integrated circuits (e.g., application-specific integrated circuits (ASICs)), or any combination thereof. In one example, some or all of the processing of data collected using an OCT probe and a processor-based system is implemented as a set of computer program instructions, which are converted into a computer-executable format, stored themselves on a computer-readable medium, and executed by a microprocessor under the control of an operating system. Thus, query responses and input data are converted into processor-understandable instructions suitable for generating imaging data, detecting lumen boundaries, detecting stent struts, comparing measured vertical distances to set thresholds, image comparison, signal processing, lumen detection, stent detection, and comparison of detected stents, as well as performing other features and examples described above in other ways.

[0052] Computer program logic that implements all or part of the functions described above can be implemented in a variety of forms, including, but not limited to, source code, computer executable, and various intermediate forms (e.g., forms generated by an assembler, compiler, linker, or locator). Source code may include a set of computer program instructions implemented in any of the programming languages ​​(e.g., object code, assembly language, or high-level languages ​​such as Fortran, C, C++, Java, or HTML) for use with various operating systems or operating environments. Source code may define and use various data structures and communication messages. Source code can be made into a computer executable form (e.g., via an interpreter), or source code can be converted into a computer executable form (e.g., via a translator, assembler, or compiler).

[0053] Computer programs can be permanently or temporarily fixed in any format (e.g., source code format, computer executable format, or intermediate format) on tangible storage media such as semiconductor memory devices (e.g., RAM, ROM, PROM, EEPROM, or flash programmable RAM), magnetic memory devices (e.g., diskettes or hard disks), optical memory devices (e.g., CD-ROMs), PC cards (e.g., PCMCIA cards), or other memory devices. Computer programs can also be fixed in any format as signals that can be transmitted to a computer using any of the various communication technologies. These communication technologies include, but are not limited to, analog, digital, optical, wireless (e.g., Bluetooth), networking, and internetworking technologies. Computer programs can also be distributed in any format as removable storage media with accompanying printed or electronic documentation (e.g., shrink-wrapped software), pre-loaded onto computer systems (e.g., system ROM or hard disks), or distributed from servers or electronic bulletin boards via communication systems (e.g., the Internet or the World Wide Web).

[0054] Hardware logic (including programmable logic used with programmable logic devices) that performs all or part of the functions described herein may be designed using conventional manual methods, or it may be designed, captured, simulated, or documented electronically using various tools such as computer-aided design (CAD), hardware description languages ​​(e.g., VHDL or AHDL), or PLD programming languages ​​(e.g., PALASM, ABEL, or CUPL).

[0055] Programmable logic can be permanently or temporarily fixed to tangible storage media such as semiconductor memory devices (e.g., RAM, ROM, PROM, EEPROM, or flash programmable RAM), magnetic memory devices (e.g., diskettes or fixed disks), optical memory devices (e.g., CD-ROMs), or other memory devices. Programmable logic can be fixed to signals that can be transmitted to a computer using any of the various communication technologies, including but not limited to analog, digital, optical, wireless (e.g., Bluetooth), networking, and internetworking technologies. Programmable logic can also be distributed as removable storage media with accompanying printed or electronic documentation (e.g., commercial software), preloaded into computer systems (e.g., system ROM or fixed disks), or delivered via communication systems (e.g., the Internet or the World Wide Web) from a server or electronic bulletin board.

[0056] Various examples of appropriate processing modules are discussed in more detail below. As used herein, a module refers to software, hardware, or firmware suitable for performing a particular data processing task or data transmission task. In some examples, a module refers to a software routine, program, or other memory-resident application suitable for receiving, transforming, routing, and processing instructions, or various types of data such as OCT scan data and other information of interest.

[0057] The computers and computer systems described herein may include operablely connected computer-readable media, such as memory for storing software applications used when acquiring, processing, storing, and / or communicating data. Such memory may be internal, external, remote, or local with respect to the operablely connected computer or computer system.

[0058] Memory may also include, for example, without limitation, any means for storing software or other instructions, including hard disks, optical disks, floppy disks, DVDs (Digital Multipurpose Disks), CDs (Compact Disks), Memory Sticks, flash memory, ROMs (Read-Only Memory), RAMs (Random Access Memory), DRAMs (Dynamic Random Access Memory), PROMs (Programmable ROMs), EEPROMs (Extended Erasable PROMs), and / or other similar computer-readable media.

[0059] In general, computer-readable memory media applicable in connection with the examples of the disclosure described herein may include any memory medium capable of storing instructions executed by a programmable device. Where applicable, the method steps described herein may be embodied or executed as instructions stored on one or more computer-readable memory media. These instructions may be software embodied in various programming languages ​​such as C++, C, Java, and / or various other types of software programming languages ​​that may be applied to create instructions according to the examples of the disclosure.

[0060] A storage medium may be non-transient or may include a non-transient device. Therefore, a non-transient storage medium or device may include a tangible device, meaning that the device may change its physical state but still possess a concrete physical form. Thus, for example, non-transient refers to the fact that the device remains tangible even if this change in state occurs.

[0061] (Example method) Figure 3 shows a method 300 for acquiring one or more calibration spectra. The calibration spectra can be used to calibrate the OCT signal derived from a sample such as sample 140 to improve resolution, signal-to-noise ratio, and suppress image sidelobe artifacts.

[0062] As used in reference to Method 300, the following notations can be used. The following equations can be derived and understood by referring to the notations. Those skilled in the art will understand that equivalent and similar steps, symbols, and notations may be used. k: Wavenumber. j: Imaginary unit in the complex domain. KL: k linearization. DC: Dispersion correction. SF: Spectrum flattening. n: Sampling index by the A / D card used. I: Intensity I P1 (n): The intensity at position P1, where n is a variable of the sampling index by the A / D card. z x : Depth position x with respect to the zero-delay plane. Px: Imaging position x on the positive plane, for example, x = 1, 2, 3... -Px: Imaging position x on the negative plane. Φ x (n): Optical phase acquired at the x position on the positive plane. Φ -x (n): Optical phase acquired at the x position on the negative plane. Φ disp : Optical dispersion phase. Φ KL (n): Non-dispersive k spectrum after the k linearization process. X d (n): Phase compensation term used to compensate for the asymmetry with respect to the zero-delay line. z d : PSF peak position after FFT, where d is Px and -Px. s0: Corrected fringe after k linearization. s1: Corrected fringe after the KL and DC processes. s2: Corrected fringe after the KL, DC, and spectrum flattening processes. s3: Corrected fringe after the KL, DC, and spectrum flattening processes without using the analytical form to reduce the computational cost.

[0063] In block 305, mirror measurements can be acquired. Mirror measurements are signals generated from the OCT system 100 when the mirror is positioned on the sample 140. The resulting measurements are time-varying amplitudes from the interferogram acquired by the A / D card. In this block, OCT signals, which may contain raw fringe data, can be acquired separately from both the negative and positive imaging planes, from the mirror or reference reflector within the sample arm of the OCT system. Background spectra can also be acquired by removing the sample. In some examples, raw fringe on both sides of the mirror arm can be acquired at a pixel depth of 250 to 300 pixels, corresponding to 25% to 30% of the Nyquist depth in a 1024-pixel system. In other examples, any two arbitrary positions can be selected for the pixel depth.

[0064] In some examples, measurements can be obtained as positions P1 to P3, as shown in Figure 1B. Information is derived from these measurements, as will be explained further. In some examples, the measurements from positions P1 and -P1 are taken at sampling index n, Φ P1 (n) and Φ -P1 It is used to extract the phase as a function of (n), and accordingly can be used to calculate the KL spectrum and the dispersion spectrum. The fringe obtained at P1 can be used to estimate the spectral flattening spectrum. In other examples, the k spectrum can be optimized based on the best polynomial fitting order detected by the algorithm using measurements taken at P1 to P3. In some examples, the positions of P1, P2, and P3 can be selected based on the desired or estimated depth of the sample to optimize the polynomial fitting over the entire depth range.

[0065] The characteristics of one or more acquired signals are shown with reference to Figure 4A. In some examples, the complex signal and its phase can be calculated from the measurements acquired at positions P1 to P3 using the Hilbert transform.

[0066] The k-linear phase also averages the fringe to obtain a clean optical phase Φ from the P1 and P2 positions. P1 (n) and Φ -P1 This can be calculated by obtaining (n). Using the interferograms of P1-P3, the interpolation performance of the KL spectrum is optimized, reducing or avoiding the influence of noise from deeper depths in the OCT signal.

[0067] In other words, I P1 (n)∝cos{k(n)z P1 +Φ disp (n) and I -P1 (n)∝cos{k(n)z -P1 -Φ disp Since it is known that (n)}, the instantaneous optical phase Φ P1 (n) and Φ -P1 (n) is calculated using the Hilbert transform, Φ P1 (n)=k(n)z P1 +Φ disp (n) and Φ -P1 (n)=k(n)z -P1 -Φ disp (n) can be obtained. In this block, background can be removed from the signal obtained in block 305.

[0068] In block 310, the dispersive-free k spectrum (k spectrum) Φ KL (n) is Φ KL (n) = 0.5 * {Φ P1 (n) + Φ -P1This can be calculated based on the relationship (n). The aspect of this block is further illustrated with reference to Figure 4B(1), where the solid line represents the dispersion-free k spectrum and the dashed line represents the ideal linearized k spectrum. The difference between the curves represents the nonlinearity of the k clock, which needs to be corrected in the next section. Using both curves, the raw fringe on the wavenumber phase domain is directly interpolated to obtain the KL-corrected fringe s0(n). In addition, to test the algorithm using an uncalibrated OPTIS system, instead of using the k clock, the OCT signal was acquired at the A / D card's internal sampling rate of 500 MHz. Figure 4B(2) shows an effective calibration to correct the nonlinear k spectrum to the linearized k spectrum after the KL process. Thus, the calibrated result is shown in Figure 12. Conventional techniques perform interpolation in the time domain, but performing interpolation in the time domain requires additional polynomial fitting to project the k-domain phase back onto the time domain (N=1~2048), and the accuracy is extremely sensitive to the order of the polynomial fitting. To save computational cost from the additional polynomial fitting that remaps the interpolation from the k domain to the time domain, an embodiment of this method can be selected to perform k-spectral interpolation directly in the wavenumber domain.

[0069] In some cases, as shown in Figure 4C, system noise across the entire fringe and increased sampling error near the Nyquist depth can easily lead to imaging sidelobe artifacts, further degrading system resolution at deeper depths (i.e., deeper signal depths have fewer sampling points per fringe cycle). Therefore, the dispersion-free k-spectrum needs to be optimized by a polynomial fitting function, and the order of the polynomial fitting can be optimized. The results, as shown in Figure 4C with enlarged windows Figures 455 and 465, demonstrate an improved PSF without sidelobe artifacts.

[0070] In block 315, the best polynomial order of the KL spectrum can be determined by using one or all of the following steps as part of the polynomial optimization process. ·KL spectrum Φ KL (n) can be obtained using polynomial fitting at either P1 or another mirror position (P2, P3, etc.). • KL resampling using various polynomial fitting orders can be applied to signals acquired at other locations such as P2 and P3. In some examples, polynomials of order 0 to 50 can be used, where 0 represents the raw k spectrum without fitting. Based on the PSF profile after k-spectral interpolation for each fitting order, the full width at half maximum (FWHM) (referred to as a1 in this block) and the sum of the point intensity distribution function (PSF) areas (referred to as a2 in this block) are calculated. The parameters a1 and a2 can be calculated for each depth location P1, P2, and P3 using any polynomial degree. Then, the parameters a1 and a2 are averaged along the depth and then normalized separately between 0 and 1. • We can find the polynomial degree that has the smallest (a1+a2). The polynomial order that produces the sharpest intensity peak in the preceding step is saved as the optimized fitting order of the saved k spectrum and can be used throughout the entire KL process at all other depths and for all other OCT signals in the real-time imaging stage. For an exemplary improvement in imaging artifacts, see the bottom row of Figures 4C and 11.

[0071] In block 320, the dispersion spectrum can be calculated. First, the raw fringe is interpolated using the k spectrum from block 310. The interferogram with the mirror at positions P1 and -P1 is given by the following relationship I P1 (n)∝cos{k(n)z P1 +Φ disp (n) and I -P1 (n)∝cos{k(n)z-P1 -Φ disp Described by (n)}. If the mirror position is not equidistant from the zero delay line, the dispersion phase Φ disp This can be achieved as follows: Φ disp (n) = 0.5 * {(Φ P1 (n) + n*X p1 )-(Φ -P1 (n) + n*X (-P1) )}, and X d =-z d +0.5*{z (P1) +z (-P1)}

[0072] For each mirror calibration pair, term X d Using this, the phase difference of the calibration mirrors resulting from their unequal positions relative to the zero delay line is compensated. Noise Φ disp (n) can be removed by fitting a polynomial function, as shown in Figure 4D, or simply filtered by a low-pass filter. The obtained Φ disp The (k) function can be saved for use as the inverse dispersion phase to be subtracted using the analysis form of the KL-corrected fringe s0(n). See Figure 6B for example results of the fringe data and OCT intensity shape after applying dispersion correction.

[0073] In block 325, the spectral flattening (SF) spectrum is calculated. SF(n) can be calculated using the envelope value of the KLDC-corrected fringe s1(n). In some examples, spectral flattening can further increase axial resolution performance by increasing the available bandwidth, as shown in Figure 6C. In one example, the SF window can be calculated as a calibration step and applied to future data using a window function (i.e., a Kaiser-Bessel window) to form a demodulation window, e.g., s2(n) = s1(n) × (Kaiser window) / SF(n). In one example, the envelope used can be derived from the signal acquired from one of the reference mirror positions from block 305.

[0074] In one example, the first step of the spectral flattening process is to calculate the envelope of the fringes by taking the absolute number of KLDC-corrected fringes after the Hilbert transform has been applied. This step can occur when the mirror is positioned within the sample arm. Once the spectral envelope is calculated, it can be fitted by a polynomial function (see Figure 4E as an example).

[0075] In block 330, the various spectra calculated in steps 310 to 325 can be stored. In this block, the various spectra can be stored or moved to different parts of the computing device 200.

[0076] Figure 4A shows various aspects of the acquired input signal. Graph 410 shows the input signal acquired from the positive imaging plane. Graph 420 shows the input signal acquired from the negative imaging plane. Graphs 410 and 420 can also be described as examples of k-clock sampled interferograms (raw fringes) from an OCT system. Graphs 410 and 420 represent empirically acquired data, but the intensity of the fringe signals in the graphs can be described by the wavenumber "k" and variance, and relation I P1 (n)∝cos{k(n)z P1 +Φ disp (n) and I -P1 (n)∝cos{k(n)z -P1 -Φ disp (n)} has the variance, which can be measured in radians or any other angular scale and represents the optical phase that determines the amplitude modulation of the sinusoidal signal.

[0077] Referring to Figure 4B, graphs 430 (Figure 4B(1)) and 440 (Figure 4B(2)) are shown. Figure 4B(1) is graph 430, showing the dispersion-free k spectrum compared to the ideal linearized k spectrum. Figure 4B(2) is graph 440, showing the effective correction for the uncalibrated k spectrum after the KL process. In Figure 4B, the dashed line may represent the optical phase calculated after the k-linearization process.

[0078] Figure 4C shows graphs 450 and 460. Graph 450 shows the intensity without k-spectral optimization. Graph 455 highlights one peak indicating heterogeneous sidelobes. Graph 460 shows the same signal with k-spectral optimization. Graph 465 highlights the same peak as 455, indicating suppression of sidelobes, and the graph is symmetrical after k-spectral optimization.

[0079] Figure 4D shows the polynomial fitting Φ disp (n) shows graph 470, which represents the graph of the saved raw distributed data.

[0080] Figure 4E shows Graph 480. Graph 480 shows example data of KLDC-corrected fringes and their envelope calculated using polynomial fitting. The calculated envelope can be saved for later use.

[0081] Figure 5 shows a method 500 for performing real-time OCT imaging using a graphics processing unit (GPU) or a field programmable gate array (FPGA) according to an aspect of this disclosure. Method 500 can also be performed in part or in whole by other computing devices, such as device 200. Method 500 can be performed in real time while the OCT system is operating, using data acquired at the aforementioned location.

[0082] In block 505, one or more spectra can be loaded or acquired. For example, one or more spectra calculated by referring to method 300, such as a k-spectrum, a dispersion spectrum, and a spectral flattening spectrum, can be loaded from memory. In this block, background data for each fringe must be removed by a high-pass filter or acquired by blocking the sample arm.

[0083] In block 510, k-linearization (KL) correction can be performed on the collected raw fringe data. In this block, the fringe data can be interpolated with a saved k-spectrum having an optimized polynomial fitting order. In some examples, block 510 can use k-spectrum data saved with reference to method 300 and shown with reference to Figure 4B. In this step, the loaded k-spectrum can be used to correct the acquired raw fringe data by performing cubic spline interpolation in wavenumber space (k-space) on the collected raw fringe data. Performing cubic spline interpolation in wavenumber space makes it possible to scale the raw fringe from the nonlinear wavenumber domain to the linear wavenumber domain before performing the FFT process. Spline interpolation is a form of piecewise polynomial interpolation that avoids the problem of overfitting that causes spike errors. In some examples, an FPGA or GPU can be used to efficiently perform cubic spline interpolation in the wavenumber domain, making it possible to perform KL correction in real time, such as during the operation of an OCT system. The raw and KLDC-corrected fringe data are shown with respect to Figure 6B. The spectrum can also be loaded into a non-uniform FFT without a numerical interpolation process.

[0084] In block 515, the variance correction (DC) is calculated by referring to method 300, and more specifically, to block 320. disp This can be performed on the KL-corrected fringe data generated in block 510 through the use of the (n) function. In this block, dispersion compensation is performed to cancel out the dispersion phase, and the acquired OCT signal is corrected. Since a polynomial fitting function or a low-pass filter is used, a more robust correction can be performed compared to the raw dispersion data.

[0085] This block allows us to perform a Hilbert transform on fringe data into an analytic format, where j is the imaginary unit, and -jΦ dispThe dispersion phase can be compensated using the exponential function of (n).

[0086] In this block, the new fringe function can be computed by taking the real part of the corrected fringe data multiplied by the complex-valued phase. s1(n)=Real{analytical form of s0(n)×exp[-jΦ disp (n)]}

[0087] The real-valued function described above is the "real" part of the fringe, which has been linearized k-wise and transformed into a complex-valued analytical form.

[0088] In block 520, spectral flattening can be performed on the results obtained in block 515. In this block, the spectral envelope can be removed. Despite both KL and DC corrections, the ramping envelope of the fringe shown in Figure 4E can still limit the axial resolution performance due to the constrained bandwidth. In this block, the real-valued envelope calculated in block 320 can be used. The spectral envelope from the calibration is incorporated into a window function (e.g., a Kaiser window) and used as a demodulation window to flatten the KLDC-corrected fringe.

[0089] In some examples, the following steps can be performed in block 520. s2(n)=s1(n)×(kaiser window(n)) / SF(n)

[0090] The spectral envelope SF(n) described above is defined as the envelope of s1(n).

[0091] Figure 4B shows Graph 430, which illustrates the generated k-spectrum and the ideal k-spectrum. Graph 620 is an enlarged portion of the original fringe versus the KL-corrected fringe according to an embodiment of Method 300. The fitted spectrum can be saved and used for real-time imaging, as described in Method 500. Graph 620 shows the original data and the corrected data after the k-linearization correction step has been performed.

[0092] Figure 6A shows Graph 620, which is the interferogram as a function of the sampling index. Graph 620 shows the original fringe signal and the KL-corrected fringe signal.

[0093] Figure 6B shows graph 630, which contains the difference between the KL-corrected fringe signals before and after applying dispersion correction. A simplified form that accelerates the real-time imaging process can be described as follows: s3(z)=abs{s0(n)×exp[-jΦ disp (n)]×(kaiser window(n)) / SF(n)}

[0094] The absolute function abs mentioned above is obtained by multiplying the "magnitude or coefficient" part of the k-linearized fringe by the dispersion phase.

[0095] Graph 640 shows the narrower PSF after KLDC processing versus k-clock linearization.

[0096] Figure 6C shows an example of the difference in bandwidth before and after applying spectral equalization in Graph 650. Note that the plot below shows a wider FWHM bandwidth and a more symmetrical structure after spectral equalization.

[0097] Figure 6D shows Graph 660 as an example illustrating FWHM at different imaging depths using different methods. Data were acquired from an OPTIS OCT system with unoptimized k-delay and no strong system dispersion. Graph 660 shows the improvement in FWHM across all imaging depths between (i) a signal using only k-clock data, (ii) a signal using only the KL technique, (iii) a signal using the KL and DC techniques, and (iv) a signal using the KL, DC, and SF techniques. A smaller FWHM indicates improved axial resolution in the OCT image.

[0098] Figure 7 shows a method 700 for generating OCT real-time imaging incorporating a KLDC configuration, according to an aspect of this disclosure.

[0099] In block 705, fringe data can be acquired from the OCT system's sensors. As one example, the system described with reference to Figure 1 can be used. In another example, the Abbott OPTIS system can be used.

[0100] In block 710, KL, DC, and spectral flattening may be performed. In this step, one or more steps relating to method 500 may be performed.

[0101] In block 715, zero padding and windowing can be performed. Windowing involves retaining a portion of the signal within a selected interval. Mathematically, windowing is equivalent to applying a window function that has zero values ​​outside the selected interval. Zero padding is a process that can add zeros to the end of a signal to extend its length. This process can also allow the acquired signal to be processed more efficiently by the Fast Fourier Transform. Zero padding can increase the sample point, allowing the reconstructed signal to approach the theoretical limit of optical resolution.

[0102] In block 720, a Fast Fourier Transform (FFT) can be performed on the acquired data. The FFT can be an algorithm that computes the Discrete Fourier Transform or its inverse.

[0103] In block 725, the data is converted to a logarithmic scale. In some examples, the FFT data can be normalized before being converted to a logarithmic scale.

[0104] In block 730, intensity images can be generated. The generated intensity images can be displayed on a monitor. The image data can be based on logarithmic scale values ​​acquired in block 725.

[0105] Figure 8 shows an exemplary OPTIS system with k-delay adjustment set according to a standard manufacturing procedure. The plot shows that the KLDCsf (KLDC and spectral flattening) process corrects imaging sidelobes and improves system resolution in commercially available systems that meet the standards.

[0106] Figure 9 shows the KLDCsf performance in an OPTIS system with k-delay adjustment within specifications. The PSF profile shows a sharper beam waist, and the system FWHM shows excellent system resolution across the entire imaging depth. Figure 9 shows intensity plots 910, 920, and 930 at various imaging depths, along with the intensity of the signal acquired using the KLDCsf technique and the intensity of the same signal without the KLDCsf technique.

[0107] Figure 9 also shows plot 940, which shows imaging depth on the x-axis and FWMH in micrometers on the y-axis. As shown in plot 940, when using the KLDCsf technique, the FWMH is consistently narrower across the entire imaging depth compared to systems using k-clocks.

[0108] Figure 10 shows plots 1010 and 1020 relating to the performance of KLDCsf on an OPTIS system without k-delay adjustment and strong dispersion (not within specifications). Plots 1010 and 1020 also demonstrate the best resolution and SNR by performing KLDCsf with an optimized k spectrum. Plot 1010 shows imaging depth on the x-axis and FWHM in micrometers on the y-axis. Plot 1010 shows a smaller FWHM with the KLDCsf process and a slight improvement in FWHM when the KLDCsf process is applied to data resampled using the k-clock. Plot 1020 shows the signal-to-noise ratio (SNR) on the y-axis and imaging depth on the x-axis, showing a higher signal-to-noise ratio using the KLDCsf technique compared to signals acquired using the k-clock.

[0109] Figure 11 shows graph 1110 of a single-line scan (A-line) imaged from a Kapton tape roll by an OPTIS system operating within standards. KLDCsf was applied to the raw data and to the data linearized in k-space using k-clocks. Graph 1110 shows pixel depth on the x-axis and intensity in decibels on the y-axis. As can be seen from the unlabeled arrows indicating peaks, the intensity peaks obtained using KLDCsf with an optimized k-spectrum are higher than the intensity peaks obtained by using KLDCsf alone at deeper depths (above 80% of the Nyquist imaging depth), resulting in better SNR and system resolution at depth. No significant difference was observed below 80% of the Nyquist depth.

[0110] Figure 12 shows the performance of KLDCsf applied to the mirror signal of the OPTIS system without using a k-clock. As can be seen from Graph 1210, KLDCsf processing of the signal yields sharper and more symmetrical peaks at sample depth, but the mirror signal cannot be reconstructed using data that is not linearized in k-space.

[0111] Figure 13 shows plot 1310 illustrating the resolution of a standard OPTIS system using its internal k-clock and the same system without a connected k-clock, using the KLDCsf process. As can be seen from 1310, KLDCsf without a k-clock improves the resolution of the original k-clock configuration.

[0112] The aspects, embodiments, features, and examples of this disclosure are intended to be illustrative in all respects and are not intended to limit the disclosure, the scope of which is defined solely by the claims. Other examples, modifications, and uses will be apparent to those skilled in the art without departing from the spirit and scope of the claimed disclosure.

[0113] The use of headings and paragraphs in this application is not intended to limit the disclosure, and each paragraph may apply to any aspect, example, or feature of the disclosure.

[0114] Throughout this application, where a component is described as having, including, or comprising certain components, or where a process is described as having, including, or comprising certain process steps, it is intended that the components of this teaching consist essentially of the listed components, and the processes of this teaching consist essentially of the listed process steps.

[0115] Where it is referred to in this application that an element or component is included in and / or selected from the enumerated list of elements or components, it should be understood that the element or component may be any one of the enumerated elements or components, or may be selected from a group of two or more of the enumerated elements or components. Furthermore, it should be understood that the elements and / or features of the constructs, apparatus, or methods described herein may be combined in various ways, whether expressly or implicitly, without departing from the spirit and scope of this teaching.

[0116] The use of the terms “include,” “includes,” “including,” “have,” “has,” or “having” should generally be understood as open-ended and non-restrictive, unless otherwise specified.

[0117] In this specification, the use of the singular includes the plural unless otherwise specified (and vice versa). Furthermore, unless the context explicitly indicates otherwise, the singular forms "a," "an," and "the" include the plural. Furthermore, where the terms "about" or "substantially" are used before a quantitative value, this instruction also includes the specific quantitative value itself unless otherwise specified. As used herein, the terms "about" or "substantially" refer to variations in quantity that may occur, for example, through measurement or handling procedures in the real world, through accidental errors in these procedures, through differences / defects in the manufacture of materials such as composite tapes, through defects, and variations that would be recognized as equivalent by a person skilled in the art, unless such variations encompass known values ​​implemented by the prior art. Typically, the terms "about" or "substantially" mean that the stated value is greater or less than 1 / 10, for example, ±10%, of the stated value or range of values.

[0118] It should be understood that the order of steps or the order in which certain actions are performed is not important as long as this instruction remains usable. Furthermore, two or more steps or actions can be performed simultaneously.

[0119] The use of headings and paragraphs in this application does not imply any limitation of the disclosure. Each paragraph may apply to any aspect, example, or feature of the disclosure. Only claims using the term “means for” are intended to be construed under 112(6) of the United States Patent Act. Where there is no enumeration of “means for” in a claim, such claim should not be construed under 112 of the United States Patent Act. No limitation from this specification is intended to be incorporated into any claim unless such limitation is expressly included in the claims.

[0120] Given a value or a range of values, each value and the endpoints of a given range, as well as the values ​​between them, may be increased or decreased by 20%, while remaining within the teachings of this disclosure unless any different range is specifically stated.

[0121] Where a range or list of values ​​is provided, each value interposing between the upper and lower limits of that range or list of values ​​is contemplated individually and is included in this disclosure as if it were specifically enumerated herein. Furthermore, the ranges between the upper and lower limits of a given range and smaller ranges containing them are contemplated and are included in this disclosure. Exemplary lists of values ​​or ranges are not intended to exclude other values ​​or ranges between the upper and lower limits of a given range and containing them.

[0122] It is understood that the figures and descriptions in this disclosure have been simplified to show elements that are appropriate for a clear understanding of this disclosure, while omitting other elements for clarity. A person skilled in the art, however, will recognize that these and other elements may be desirable. However, since such elements are well known in the art and do not facilitate a better understanding of this disclosure, no description of such elements is provided herein. It should be understood that the figures are presented for illustrative purposes only, and not as structural diagrams. Omitted details and modifications or alternative examples are within the knowledge of a person skilled in the art.

[0123] In certain embodiments of this disclosure, it may be recognized that a single component can be replaced by multiple components, and multiple components can be replaced by a single component, in order to provide an element or structure, or to perform one or more given functions. Such replacements are considered to be within the scope of this disclosure unless they are not available for performing a particular embodiment of this disclosure.

[0124] The examples presented herein are intended to illustrate possible and specific embodiments of the Disclosure. It can be recognized that these examples are intended primarily for the benefit of those skilled in the art. Variations of these figures or the operations described herein may exist without departing from the spirit of the Disclosure. For example, in certain particular cases, method steps or operations may be performed or carried out in a different order, or operations may be added, deleted, or modified.

[0125] The disclosed aspects of the technology may include any combination of the following features:

[0126] A method for processing optical signals to improve the resolution of optically acquired images includes calibrating an optical system, the calibration comprising: one or more processors acquiring at least a first mirror measurement and a second mirror measurement, wherein the first and second mirror measurements are collected from both sides of a zero delay line, and each mirror measurement is an interferogram or signal or system impulse response; one or more processors acquiring the amplitude and phase of the first and second mirror measurements; resampling at least one of the first and second mirror measurements to be (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time to generate a resampled mirror measurement; and calculating an optimized fitting of the resampled mirror measurement using a function. Any number of mirror measurements can be used for calibration. For example, a few, tens, hundreds, or more mirror measurements can be acquired and used for calibration.

[0127] The function used for fitting may be one of a polynomial function, a cubic spline fitting function, a radial basis function, or a piecewise function. The method may further include the steps of performing k-linearization on the raw interferogram of Miller measurements, the Miller measurements being acquired with or without k-clocks, and generating a resampled interferogram, the resampled Miller measurements being in k-space.

[0128] k-linearization can be performed using an interpolation function, which is one of the following in k-space: cubic spline interpolation, cubic interpolation, or linear interpolation.

[0129] The method may further include the step of performing spectral flattening for at least one Miller measurement by using an envelope calculated from a specific Miller measurement.

[0130] Each Miller measurement can represent the system impulse response.

[0131] The method may further include the steps of calculating variance compensation for at least two Miller measurements and optionally compensating for the variance. The method may further include the step of saving the spectral envelope calculated during spectral flattening.

[0132] In some examples, the optimized polynomial is determined based on at least two Miller measurements. In some examples, three or more Miller measurements may be used. The optimization of the optimized polynomial can be based on a polynomial fitting order 1 to n, where n is a fixed integer. The method may further include the step of finding the polynomial fitting order that minimizes (i) the area of ​​the point image intensity distribution function, (ii) the full width at half maximum (FWHM) resolution, or a combination of (i) and (ii) for each Miller measurement.

[0133] (i) The method may further include the step of saving calibration criteria related to k-linearization (KL), (ii) variance correction (DC), and (iii) spectral flattening (SF).

[0134] The method may further include the step of receiving an OCT signal or interference signal from a sample. The method may further include the step of generating a new fringe or resampling the interferogram after linear k interpolation to k-linearize the received OCT signal or interference signal using a KL calibration standard. The method may further include the step of performing variance correction on the k-linearized OCT signal or interference signal using a DC calibration standard. The method may further include the step of removing the envelope of the k-linearized and variance-corrected OCT signal or interference signal. The method may further include the step of converting the OCT signal or interference signal into an OCT image to be displayed on a display.

[0135] A method for generating an OCT image from an OCT signal may include the steps of: acquiring an OCT signal corresponding to a sample and loading a calibration spectrum; wavenumber linearization (KL) of the OCT signal to generate a KL OCT signal; variance correction (DC) of the KL OCT signal to generate a KL DC OCT signal; spectral flattening (sf) of the KL DC OCT signal to generate a final KLDCsf OCT signal; and post-processing of the final OCT signal to generate an OCT image, wherein the calibration spectra used to generate the KL OCT signal, KL DC OCT signal, and final KLDCsf OCT signal are based on a calibration criterion generated in the calibration step.

[0136] The OCT signal can be, for example, an interferogram.

[0137] A calibration criterion can be generated during the calibration phase using at least a first and a second mirror measurement, the first and second mirror measurements being collected from both sides of a zero-delay line, and each mirror measurement being an interferogram or a signal or system impulse response. According to some examples, additional mirror measurements may be used.

[0138] A system for displaying a set of images of a subject comprises a memory for storing image data and time-varying data corresponding to the subject, and one or more processors communicating with the memory, wherein one or more processors acquire at least a first and second Mirror measurement, the first and second Mirror measurements being collected from both sides of a zero-delay line, each Mirror measurement being an interferogram or a signal or system impulse response, and one or more processors are operable to calculate the amplitude and phase of the first and second Mirror measurements, to resample at least one of the first and second Mirror measurements to be at least one of (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time, to generate a resampled Mirror measurement, and to use a function to calculate an optimized fitting of the resampled Mirror measurement.

[0139] The system or computer-readable medium may be configured to implement any combination of the above features.

[0140] A system for performing optical coherence tomography on a sample may include a light source, a reference mirror with a variable position, a display, a memory for storing image data corresponding to the sample, and one or more processors communicating with the memory, wherein each time the system is started, the one or more processors are operable to acquire at least one calibration parameter from mirror measurements, acquire a signal from the sample, apply at least one calibration parameter to the acquired signal, wherein the calibration parameter is at least one of wavenumber linearization, dispersion correction, or spectral flattening parameters, and generate an image from the corrected acquired signal.

[0141] The system may further include an optical switch and a calibration mirror configured to generate a calibration signal.

[0142] A computer-readable medium may include instructions executable by one or more processors to perform a method comprising acquiring a first and second mirror measurement, the first and second mirror measurements being collected from both sides of a zero-delay line, each mirror measurement being an interferogram or a signal or system impulse response, calculating the amplitude and phase of the first and second mirror measurements, resampling at least one of the first and second mirror measurements to be at least one of (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time to generate a resampled mirror measurement, and using a function to calculate an optimized fitting of the resampled mirror measurement.

[0143] A computer-readable medium can store instructions that can be executed by one or more processors to perform a method each time the system is started, which includes acquiring at least one calibration parameter, acquiring a signal from a sample, applying at least one calibration parameter to the acquired signal (where the calibration parameter is at least one of wavenumber linearization, dispersion correction, or spectral flattening parameters), and generating an image from the corrected acquired signal.

[0144] A method for processing optical signals to improve the resolution of an optically acquired image may include calibrating an optical system, the calibration comprising: one or more processors acquiring a first mirror measurement and a second mirror measurement, wherein the first and second mirror measurements are collected from both sides of a zero delay line, and each mirror measurement is an interferogram or signal or system impulse response; one or more processors acquiring the amplitude and phase of the first and second mirror measurements; calculating variance compensation for at least one mirror measurement and optionally compensating for variance; and calculating an optimized fitting for the mirror measurements using a function. Furthermore, in order to maintain the disclosures made at the time of filing this application, the contents of claims 1 to 20 at the time of filing this application are added below. (Claim 1) A method for processing optical signals, including calibrating an optical system, The aforementioned calibration is The steps include: one or more processors acquiring at least a first mirror measurement and a second mirror measurement, wherein the first mirror measurement and the second mirror measurement are collected from both sides of a zero delay line; The steps include: one or more processors acquiring the amplitude and phase of the first mirror measurement and the second mirror measurement; The steps of generating a resampled Miller measurement by resampling at least one of the first Miller measurement and the second Miller measurement so that it is (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time, The steps include: using a function to calculate an optimized fitting for the resampled Miller measurements; A method that includes [something]. (Claim 2) The method according to claim 1, wherein the function is one of a polynomial function, a cubic spline fitting function, a radial basis function, or a piecewise function. (Claim 3) The method according to claim 1, comprising the steps of performing k-linearization on the Miller measurement, wherein the Miller measurement is acquired with or without using a k-clock, and generating a resampled interferogram, wherein the resampled Miller measurement is in k-space. (Claim 4) The method according to claim 3, wherein the k-linearization is performed using an interpolation function, the interpolation function being one of cubic spline interpolation, cubic interpolation, or linear interpolation in k-space. (Claim 5) The method according to claim 1, further comprising the step of performing spectral flattening for at least one Miller measurement by using an envelope calculated from specific Miller measurements. (Claim 6) The method according to claim 1, wherein each Miller measurement is one of an interferogram, a signal, or a system impulse response. (Claim 7) The method according to claim 1, further comprising the step of calculating variance compensation for at least two Miller measurements. (Claim 8) The method according to claim 5, further comprising the step of saving the spectral envelope calculated during spectral flattening. (Claim 9) The method according to claim 1, further comprising the step of determining the order of a polynomial fitting that minimizes the sum of at least one of (i) the area of ​​the point image intensity distribution function, or (ii) the full width at half maximum (FWHM) resolution. (Claim 10) The method according to claim 1, further comprising the step of saving calibration criteria related to (i) k-linearization (KL), (ii) variance correction (DC), and (iii) spectral flattening (SF). (Claim 11) The method according to claim 1, further comprising the step of receiving an OCT signal or interference signal from a sample. (Claim 12) The method according to claim 11, further comprising the step of generating a new fringe or resampling an interferogram after linear k interpolation to linearize the received OCT signal or interference signal using the KL calibration criterion. (Claim 13) The method according to claim 12, further comprising the step of performing dispersion correction on the k-linearized OCT signal or interference signal using the DC calibration reference. (Claim 14) The method according to claim 13, further comprising the step of removing the envelope of the k-linearized and variance-corrected OCT signal or interference signal. (Claim 15) The method according to claim 11, further comprising the step of converting the OCT signal or interference signal into an OCT image to be displayed on a display. (Claim 16) A memory that stores image data and time-varying data corresponding to the subject, One or more processors that communicate with the aforementioned memory A system that is equipped with, The one or more processors described above are: The one or more processors acquire at least a first and a second mirror measurement, the first and second mirror measurements being collected from both sides of a zero-delay line, and each mirror measurement being an interferogram or a signal or system impulse response. The one or more processors calculate the amplitude and phase of the first mirror measurement and the second mirror measurement, At least one of the first Miller measurement and the second Miller measurement is resampled to (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time, thereby generating a resampled Miller measurement. The function is used to calculate the optimized fitting of the resampled Miller measurements. A system capable of operating in this manner. (Claim 17) The system according to claim 16, wherein the function used for fitting is one of a polynomial function, a cubic spline fitting function, a radial basis function, or a piecewise function. (Claim 18) The one or more processors perform k-linearization on the raw interferogram mirror measurements, and the mirror measurements are obtained with or without using k clocks. The system according to claim 16, which is further operable to generate a resampled interferogram, wherein the resampled mirror measurements are in k-space. (Claim 19) The system according to claim 16, wherein one or more processors are further operable to perform spectral flattening for at least one Miller measurement by using an envelope calculated from a specific Miller measurement. (Claim 20) A computer-readable medium containing instructions that can be executed by one or more processors to carry out a method, wherein the instructions are At least a first and a second mirror measurement are obtained, the first and second mirror measurements being collected from both sides of the zero delay line, and each mirror measurement is an interferogram or a signal or system impulse response. The amplitude and phase of the first mirror measurement and the second mirror measurement are calculated. At least one of the first Miller measurement and the second Miller measurement is resampled to (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time, thereby generating a resampled Miller measurement. A function is used to calculate the optimized fitting of the resampled Miller measurements. A computer-readable medium that includes the following:

Claims

1. A method for processing optical signals, including calibrating an optical system, The aforementioned calibration is A step in which one or more processors acquire at least a first mirror measurement and a second mirror measurement, wherein the first mirror measurement and the second mirror measurement are collected from both sides of the zero delay line, The steps include: one or more processors acquiring the amplitude and phase of the first mirror measurement and the second mirror measurement; The steps of generating a resampled Miller measurement by resampling at least one of the first Miller measurement and the second Miller measurement so that it is (i) linear wavenumber (k) space, (ii) linear wavelength, and (iii) linear time, The steps include performing spectral equalization for at least one Miller measurement by using an envelope calculated from a specific Miller measurement and saving the spectral envelope calculated during spectral equalization, The steps include: using a function to calculate an optimized fitting for the resampled Miller measurements; A method that includes [something].

2. The method according to claim 1, wherein the function is one of a polynomial function, a cubic spline fitting function, a radial basis function, or a piecewise function.

3. The method according to claim 1, comprising the steps of performing k-linearization on the Miller measurement, the Miller measurement being acquired with or without using a k-clock, and generating a resampled interferogram, wherein the resampled Miller measurement is in k-space.

4. The method according to claim 3, wherein the k-linearization is performed using an interpolation function, the interpolation function being one of cubic spline interpolation, cubic interpolation, or linear interpolation in k-space.

5. The method according to claim 1, wherein each Miller measurement is one of an interferogram, a signal, or a system impulse response.

6. The method according to claim 1, further comprising the step of calculating variance compensation for at least two Miller measurements.

7. The method according to claim 1, further comprising the step of determining the order of a polynomial fitting that minimizes the sum of at least one of (i) the area of ​​the point image intensity distribution function, or (ii) the full width at half maximum (FWHM) resolution.

8. The method according to claim 1, further comprising the step of saving calibration criteria related to (i) k-linearization (KL), (ii) variance correction (DC), and (iii) spectral flattening (SF).

9. The method according to claim 1, further comprising the step of receiving an OCT signal or interference signal from a sample.

10. The method according to claim 9, further comprising the step of generating a new fringe or resampling an interferogram after linear k interpolation to linearize the received OCT signal or interference signal by using a KL calibration standard.

11. The method according to claim 10, further comprising the steps of performing dispersion correction on the k-linearized OCT signal or interference signal using a DC calibration reference, and removing the envelope of the k-linearized and dispersion-corrected OCT signal or interference signal.

12. The method according to claim 9, further comprising the step of converting the OCT signal or interference signal into an OCT image to be displayed on a display.

13. A memory that stores image data and time-varying data corresponding to the subject, One or more processors that are operable to perform the method according to any one of claims 1 to 12 and communicate with the memory A system equipped with [this feature].

Images