Optical coherence tomography system and method

Abstract

An optical coherence tomography system that quickly traverses its respective scan pattern and typically completes the entire two-dimensional frame faster than a conventional grating scanner completes one grating segment. To quickly traverse the scan pattern, the system performs fewer A-scans per unit scan pattern length than conventional OCT scanners. To compensate for the sparsity of sampling points along the respective scan line segments and gaps between the respective line segments of the compensation trajectory, the system acquires and combines several partially overlapping frames for each application.

Description

[0001] Citation of Related Applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 374,991, filed on September 8, 2022; U.S. Provisional Application No. 63 / 410,155, filed on September 26, 2022; U.S. Provisional Application No. 63 / 378,482, filed on October 5, 2022; U.S. Provisional Application No. 63 / 380,161, filed on October 19, 2022; U.S. Provisional Application No. 63 / 383,858, filed on November 15, 2022; and U.S. Provisional Application No. 63 / 499,210, filed on April 28, 2023, the entire contents of which are incorporated herein by reference. Background Art

[0003] Dental caries is a common disease that affects over 90% of American adults. Despite progress in preventive measures, dental caries remains the leading cause of tooth decay that requires invasive treatment to repair teeth. Over 35% of Americans do not see a dentist every year, and the Centers for Disease Control and Prevention (CDC) states that approximately 28% of people have untreated cavities. Among patients who see a dentist, dental service organizations (DSOs) state that the proportion of patients receiving an ideal dental treatment plan is only approximately 30%, and point out that the main reasons for this low acceptance rate are: high cost of care, inconvenience caused by multiple and time-consuming dental appointments, and low acceptance of cases by patients and insurance companies.

[0004] Avoiding dentists for these reasons often leads to the worsening of dental diseases, periodontal diseases, and other oral problems, such as the inability to detect oral cancer, which are associated with many adverse medical consequences, including eating disorders, speech difficulties, poor social interaction, decreased employment potential, and an increased risk of systemic diseases such as diabetes, cardiovascular diseases (such as stroke and heart disease), and Alzheimer's disease. Health problems caused by poor oral health have led to productivity losses of over $45 billion in the United States and school time losses for over 34 million young people. Therefore, there is an urgent need to meet the demand for affordable and effective dental health care.

[0005] To address these problems and increase access to dental care, a means is needed to reduce treatment costs, shorten appointment cycles, and increase patient and insurance company acceptance of cases. Treatment costs can be reduced by improving prevention and reducing the cost of restorative interventions, and appointment cycles can be shortened. Increasing early and accurate diagnosis will improve preventive care. Automation of tooth preparation or restorative treatment will shorten dental treatment time and reduce associated costs and appointment cycles.

[0006] In the early stages of dental caries, when there is sufficient calcium, phosphate, and fluoride ions in the oral cavity, the mineral loss in teeth can be reversed. These ions contribute to the remineralization of teeth. Early and accurate diagnosis of dental caries can reduce the cost of dental treatment because it allows for non-invasive treatment methods to prevent or halt the onset and progression of the disease. By using robotics to automate the labor-intensive tasks of dental restorative treatment, the treatment cost can be reduced, and the appointment cycle for dental diseases that have progressed to the remineralization stage can be shortened. However, this approach requires an improved imaging modality that exceeds the capabilities of radiographs, providing both a true tooth geometry and high sensitivity and specificity to guide the robot. Neither dental radiographs nor cone beam computed tomography (CBCT) are precise enough to replace the intraoral scanner (IOS) used in restorative dentistry. This is evidenced by the fact that dentists must use real-time visual and tactile feedback during tooth preparation to locate and remove all carious lesions.

[0007] To improve patient and insurance company acceptance of cases, a more sensitive and specific imaging modality that is easy for patients and insurance companies to understand is needed. Currently, patients are not accustomed to interpreting two-dimensional (2D) radiographs and thus cannot independently verify the need for care without provider explanation. According to DSO, three-dimensional (3D) radiographs such as CBCT avoid this problem and improve overall case acceptance. Insurance companies verify the need for care based on multiple inputs, including clinical notes and radiographs. However, radiographs have their own inherent limitations, including low sensitivity and specificity and the inability to image soft tissues and cracks in teeth. This low sensitivity and specificity of radiographs often lead to discrepancies between providers and payers, resulting in patients being uninsured and discrepancies between providers, which reduces professional trust and thus affects case acceptance.

[0008] Optical Coherence Tomography (OCT) is an excellent imaging candidate because it has several advantages over radiography in dental applications. Optical Coherence Tomography (OCT) uses near-infrared light to take 3D images of biological tissues. Compared with X-rays and computed tomography, OCT uses non-ionizing radiation and generally has a lower power level, so it may be safer for operators and patients during surgery. Exemplary advantages of OCT include fast 3D imaging, non-ionizing radiation, high dental sensitivity and clarity, and high spatial resolution (currently about 1μm to 20μm). However, OCT also has some limitations that restrict its application in dentistry, such as limited penetration depth, a small field of view (FOV) that hinders full-arch imaging, a long acquisition time that may cause motion distortion within a single volume, and the need for complex registration to achieve the surface fidelity required for an intraoral scanner (IOS) or to guide automated tooth preparation surgery. Summary of the Invention

[0009] The present disclosure provides improved Optical Coherence Tomography (OCT) systems and methods. In some embodiments of the present disclosure, a tomography system is provided. The system may include a probe housing, an optical coherence tomography system, a movable mirror, a motor, and a controller. The probe housing defines a window and is configured to translate along a path adjacent to an anatomical structure within a living patient. The path is not necessarily straight. The anatomical structure has a surface.

[0010] The optical coherence tomography system includes a sample arm and an optical detector. A portion of the sample arm extends through the window into free space outside the probe housing.

[0011] The movable mirror system is disposed within the probe housing and is configured to redirect the sample arm. The motor is disposed within the probe housing and is coupled to the mirror system. The controller is configured to automatically drive the motor to repeatedly change the orientation of the mirror system relative to two different axes, thereby repeatedly scanning the surface of the anatomical structure along a trajectory according to a determined, smooth two-dimensional scan pattern with the light of the sample arm.

[0012] The trajectory is relative to the housing. Thus, even though different parts of the anatomical structure can be scanned by successive scans, the trajectory does not change with the scans. Traversal refers to the complete repetition of the scan pattern, although the scan pattern may not form a closed loop.

[0013] Successive means one after another without interruption, but does not necessarily include the first (traversal).

[0014] The anatomical structure may have a three-dimensional surface, but when looking down along the axis of the scanner towards the structure, the light of the sample arm is scanned in two dimensions (not counting the depth scan provided by OCT).

[0015] A smooth function has a uniquely defined first derivative (slope or gradient) at each point. Graphically, a smooth function of a single variable can be plotted as a single continuous line without sudden bends or breaks. In some embodiments, the scan pattern is at least CO smooth. (See https: / / en.wikipedia.org / wiki / Smoothness)

[0016] Scanning refers to a complete traversal of a two-dimensional scan pattern, although this traversal may not form a closed loop, for example due to movement of the housing during scanning.

[0017] Each traversal of the scan pattern defines an outer boundary of a corresponding two-dimensional scan region on a corresponding portion of the surface of the anatomical structure. Each traversal of the scan pattern defines a plurality of gaps between corresponding line segments of the trajectory. These gaps are not illuminated by light during that traversal. In successive traversals of the scan pattern, corresponding scans are performed from different positions along the path, and the outer boundary of the corresponding scan region of each such scan partially overlaps the outer boundary of the scan region of at least one other such scan, illuminating a corresponding portion of at least some of the gaps defined by at least one other such traversal of the scan pattern.

[0018] For each traversal, the controller is configured to receive pixel image data from the optical detector. The pixel image data contains information about the corresponding portion of the surface of the anatomical structure. The pixel image data for each traversal contains a first number of pixels.

[0019] For successive traversals, corresponding scans are performed from different positions along the path, and the outer boundary of the corresponding scan region of each such scan partially overlaps the outer boundary of the scan region of at least one other such scan. The controller is configured to stitch together the pixel image data of the successive traversals to generate a stitched surface image having a number of pixels greater than the first number of pixels.

[0020] In any embodiment, the controller may be configured to analyze image features of the pixel image data of successive traversals to estimate a corresponding displacement of one outer boundary of the corresponding scan region from another outer boundary of the corresponding scan region.

[0021] The displacement is represented by a vector, which is essentially the magnitude and direction by which the outer boundaries (bounding boxes) of two successive scans do not overlap each other. This vector represents the movement of the scan wand between two scans.

[0022] In any embodiment, the scan pattern may include Lissajous figures. In any embodiment, the scan pattern may be anisotropic.

[0023] In any embodiment, the controller can be configured to receive voxel subsurface data from the optical detector for each traversal. The voxel subsurface data includes information about a corresponding subsurface portion of the anatomical structure. The voxel subsurface data for each traversal contains a second number of voxels. The controller can be configured to stitch together the voxel subsurface data for multiple consecutive traversals to generate a stitched subsurface three-dimensional volume image of voxels having a voxel count greater than the second number.

[0024] In any embodiment, the controller can be configured to use pixel image data to detect tooth surfaces of the anatomical structure (e.g., enamel / air boundary, air / dentin boundary, air / cementum boundary, as well as air / gum boundary, air / caries boundary, enamel / dentin boundary, dentin / pulp boundary, restoration / tooth boundary, gum / tooth boundary, and air / bone boundary, etc.). The controller can be configured to estimate the amount of refraction of light at the enamel / air boundary based on the difference between the refractive index of enamel and the refractive index of air. The controller can be configured to change the coordinates of the voxel subsurface data to at least partially compensate for the refraction of light at the enamel / air boundary.

[0025] In any embodiment, the controller can be configured to use voxel subsurface data to detect the enamel surface of the anatomical structure. The controller can be configured to use voxel subsurface data to detect the enamel / dentin boundary within the anatomical structure. The controller can be configured to estimate the thickness of the enamel between the enamel surface and the enamel / dentin boundary. The controller can be configured to estimate the amount of refraction of light within the enamel based on (a) the thickness of the enamel and (b) a predetermined refractive index of the enamel. The controller can be configured to change the coordinates of the voxel subsurface data to at least partially compensate for the refraction of light within the enamel.

[0026] In any embodiment, the mirror system can have a resonance frequency. The controller can be configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 50% of the resonance frequency.

[0027] In any embodiment, the controller can be configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 30% of the resonance frequency.

[0028] In any embodiment, the controller can be configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 20% of the resonance frequency.

[0029] Another embodiment of the present invention provides a tomographic scanning system. The system includes a probe housing that defines a window and is configured to translate along a path proximate to an anatomical structure within a living patient, the anatomical structure having a surface. An optical coherence tomography system includes a sample arm and an optical detector, wherein a portion of the sample arm extends through the window into free space outside the probe housing. A movable mirror system is disposed within the probe housing and is configured to redirect the sample arm. A motor is disposed within the probe housing and is coupled to the mirror system.

[0030] The controller is configured to automatically drive the motor to repeatedly change the orientation of the mirror system relative to two different axes so as to repeatedly scan the surface of the anatomical structure along a trajectory according to a determined two-dimensional scan pattern with light from the sample arm. Each traversal of the scan pattern defines an outer boundary of a corresponding two-dimensional scan region on a corresponding portion of the surface of the anatomical structure. Each traversal of the scan pattern defines a plurality of gaps between corresponding line segments of the trajectory, wherein these gaps are not illuminated by light during that traversal. In successive traversals of the scan pattern, corresponding scans are performed from different positions along the path, and the outer boundary of the corresponding scan region of each such scan partially overlaps the outer boundary of the scan region of at least one other such scan, illuminating a corresponding portion of at least some of the gaps defined by at least one other such traversal of the scan pattern.

[0031] For each traversal, the controller is configured to receive pixel image data regarding a corresponding portion of the surface of the anatomical structure from the optical detector, wherein the pixel image data for each traversal includes a first number of pixels. The controller is further configured to perform corresponding scans from different positions along the path for a plurality of successive traversals, and the outer boundary of the corresponding scan region of each such scan partially overlaps the outer boundary of the scan region of at least one other such scan to stitch together the pixel image data of the plurality of successive traversals, thereby generating a stitched surface image having a number of pixels greater than the first number of pixels.

[0032] In any embodiment, the scan pattern may include a grating.

[0033] Yet another embodiment of the present invention provides a tomographic scanning system. The system includes a probe housing that defines a window and is configured to translate along a path proximate to an anatomical structure within a living patient. The anatomical structure has a surface.

[0034] An optical coherence tomography system includes a sample arm and an optical detector, wherein a portion of the sample arm extends through the window into free space outside the probe housing. A movable mirror system is disposed within the probe housing and is configured to redirect the sample arm. A motor is disposed within the probe housing and is coupled to the mirror system.

[0035] The controller is configured to automatically drive the motor to repeatedly change the orientation of the mirror system relative to two different axes, so as to use the light of the sample arm to repeatedly scan the surface of the anatomical structure along a trajectory according to a scanning pattern, wherein at least 80% of the scanning pattern along the trajectory is smooth (continuous, unbroken). Each traversal of the scanning pattern defines the outer boundary of a corresponding two-dimensional scanning region on a corresponding portion of the surface of the anatomical structure. Each traversal of the scanning pattern defines a plurality of gaps between corresponding line segments of the trajectory, and during this traversal, these gaps are not illuminated by light. In consecutive traversals of the scanning pattern, corresponding scans are performed from different positions along the path, and the outer boundary of the corresponding scanning region of each such scan partially overlaps with the outer boundary of the scanning region of at least one other such scan, illuminating a corresponding portion of at least some of the gaps defined by at least one other such traversal of the scanning pattern.

[0036] For each traversal, the controller is configured to receive pixel image data regarding a corresponding portion of the surface of the anatomical structure from the optical detector, wherein the pixel image data for each traversal contains a first number of pixels. The controller is configured to, for a plurality of consecutive traversals, perform corresponding scans from different positions along the path, and the outer boundary of the corresponding scanning region of each such scan partially overlaps with the outer boundary of the scanning region of at least one other such scan, so as to stitch together the pixel image data of the plurality of consecutive traversals, thereby generating a stitched surface image with a pixel number greater than the first number of pixels.

[0037] In any embodiment, the scanning pattern may include a helix.

[0038] In any embodiment, the mirror may include a first mirror and a second mirror. The motor may be configured to continuously change the orientation of the first mirror along a first axis and continuously change the orientation of the second mirror along a second axis different from the first axis.

[0039] In any embodiment, the mirror and the motor may together include a biaxial microelectromechanical system.

[0040] Any embodiment may include a memory storing calibration data characterizing the optical non-ideality of each tomographic system. The controller may be configured to modify the data received from the optical detector to at least partially compensate for the optical non-ideality.

[0041] In any suitable embodiment, the optical non-ideality may include at least one of the following: lens aberration, mirror deformation caused by motor drive, and optical misalignment.

[0042] In any embodiment, the controller can be configured to drive the motor to change the orientation of the mirror system along two axes, thereby repeatedly scanning the structure along a first closed-loop two-dimensional scan trajectory and a second closed-loop two-dimensional scan trajectory, wherein the first closed-loop two-dimensional scan trajectory provides more sampling points than the second closed-loop two-dimensional scan trajectory.

[0043] Any embodiment can include a motion detector that is mechanically coupled to the probe housing and configured to detect the motion of the probe housing. The controller can be configured to automatically detect whether the motion of the probe housing is less than a predetermined value. The controller can be configured to automatically control the motor such that: when the motion of the probe housing is less than the predetermined value, the motor changes the orientation of the mirror to scan the teeth along the first closed-loop two-dimensional scan trajectory; and when the motion of the probe housing is not less than the predetermined value, the motor changes the orientation of the mirror to scan the teeth along the second closed-loop two-dimensional scan trajectory.

[0044] Optionally, the controller can be configured to warn the operator before starting a dense scan, switch between dense and sparse scans in response to user input, and / or perform dense scans periodically between sparse scans (such as performing a dense scan after every 50 sparse scans).

[0045] Any embodiment can include a pulse-to-continuous-wave optical buffer. The buffer includes a laser configured to output a series of pulses. Each pulse has a pulse width. An N-way optical beam splitter is coupled to the output of the laser, where N > 1. The buffer includes at least N - 1 delay lines. The respective inputs of each of the N - 1 delay lines are coupled to the respective outputs of the N-way beam splitter. The buffer includes an N-way optical combiner. The respective outputs of each delay line are coupled to the respective inputs of the N-way optical combiner. Each delay line is configured to impart a delay equal to a different integer multiple of the pulse width plus a constant k (k ≥ 0).

[0046] Optionally, in any embodiment having a pulse-to-continuous-wave optical buffer, the laser can be configured to output light according to a duty cycle (D), where N = (1 / D) - 1. Optionally, in any embodiment having a pulse-to-continuous-wave optical buffer, the laser can be configured to output light according to a duty cycle (D), where N = 1 / D. Optionally, any embodiment having a pulse-to-continuous-wave optical buffer can include a polarization detector optically coupled to the output of the N-way optical combiner. A polarization controller is optically coupled between the laser and one input of the N-way optical combiner. The polarization controller is communicatively coupled to the polarization detector. The polarization controller is configured to adjust the polarization of the light passing therethrough to match the polarization of the light delivered to the other input of the N-way optical combiner.

[0047] In some embodiments, provided herein is an optical coherence tomography system for dental imaging. The optical coherence tomography system includes a laser source having a sufficient A-scan rate to reduce motion artifacts and providing a synchronization signal for each OCT A-scan. In some embodiments, the system includes a beam steering subsystem for providing a synchronized scan pattern of the scan beam. In some embodiments, the system includes optics for providing sufficient optical performance for intraoral imaging of teeth. In some embodiments, the system includes a motion tracking subsystem for synchronously tracking the relative motion between the beam steering subsystem and the object to be imaged. In some embodiments, the system includes a handheld probe that encloses at least a portion of the beam steering subsystem, the optics, and the motion tracking subsystem. In some embodiments, the system includes a digitizer that sufficiently samples the OCT signal, the bandwidth of which is determined by the laser source and the optics.

[0048] In some embodiments, provided herein is a method for reconstructing 3D optical coherence tomography image data of a tooth, including a beam steering subsystem providing a known scan pattern of a scan beam and a motion tracking subsystem providing / estimating the relative motion between the beam steering subsystem and the object to be imaged. In some embodiments, the method includes the step of combining the scan pattern and the relative motion to correct motion artifacts within the scan volume. In some embodiments, the method includes the step of calculating the spatial relationship between consecutive scan volumes. In some embodiments, the method includes the step of calculating the true optical path of the imaging beam affected by different refractive indices in the scan field. In some embodiments, the method includes the step of stitching together multiple scan volumes.

[0049] In some embodiments, provided herein is an optical coherence tomography system for scanning a sample and generating imaging data of an object. The system includes: a swept-source laser configured to generate an output laser beam having a range of optical wavelengths; a beam splitter configured to receive the laser beam from the swept-source laser and split the laser beam into a sample arm and a reference arm; a handheld probe configured to be inserted into a patient's mouth, the probe enclosing a collimating element configured to receive the laser beam from the beam splitter, a beam steering subsystem including a plurality of mirrors configured to change the direction of the laser beam, and a focusing element configured to receive the laser beam from the beam steering subsystem; an optical camera positioned within the optical focus of the object and configured to be collinear with the imaging path of the laser beam and record the optical data of the object; and a digitizer configured to receive analog image data from the sample arm and the reference arm and convert the image data into digitized image data.

[0050] In some embodiments, the laser is configured to operate at a sweep duty cycle from about 35% to about 55%. In some embodiments, the laser is configured to operate at a sweep duty cycle from about 35% to about 75%. In some embodiments, the laser is configured to operate at a sweep duty cycle of about 75%. In some embodiments, the system corrects for movement of the handheld probe by providing an A-scan rate of at least 100,000 Hz and providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the system corrects for movement of the handheld probe by providing an A-scan rate of at least 150,000 Hz and providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the system corrects for movement of the handheld probe by providing an A-scan rate of at least 200,000 Hz and providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the system corrects for movement of the handheld probe by providing an A-scan rate of at least 100,000 Hz or 200,000 Hz and providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the synchronization signal for each A-scan includes the time at which each A-scan begins. In some embodiments, the reference signal includes data showing the change in the scan path of the laser relative to time. In some embodiments, the reference signal includes a Mach-Zehnder signal, an interferometer reference signal, a K-clock signal, or a combination thereof. In some embodiments, the beam steering subsystem including multiple mirrors includes a fast scanning subsystem. In some embodiments, the fast scanning subsystem includes a resonant scanner. In some embodiments, the multiple mirrors include a galvanometer mirror system. In some embodiments, the multiple mirrors include a polygonal mirror scanner. In some embodiments, the multiple mirrors include a microelectromechanical system micro-mirror scanner (MEMS). In some embodiments, the scan speed of the system is in the range of 100 Hz to 20,000 Hz, or 1500 Hz to 10,000 Hz.

[0051] In some embodiments, the system corrects the movement of the handheld probe by providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the system corrects the movement of the handheld probe by providing a synchronized scan pattern of the laser beam. In some embodiments, the system corrects the movement of the handheld probe by providing the known time and known position of multiple mirrors at a given point of the reflected laser beam. In some embodiments, the synchronized scan pattern includes a raster scan pattern. In some embodiments, the system corrects the movement of the handheld probe by determining the relative movement between the handheld probe and the optical camera. In some embodiments, the system corrects the movement of the handheld probe by correcting the digitized image data to address the relative movement of the handheld probe when generating the image data of the target. In some embodiments, correcting the digitized image data to address the relative movement of the handheld probe includes applying a block matching algorithm, a contour tracking algorithm, cross-correlation, an optical flow algorithm, a feature-based algorithm, a morphological iterative closest point algorithm, a normal distribution transform algorithm, a coherent point drift algorithm, or a combination thereof.

[0052] In some embodiments, the system is configured to operate at a sweep duty cycle of approximately 50%. In some embodiments, the system further includes a buffer system configured to increase the duty cycle of the laser, the buffer system including: a first delay module configured to receive a first portion of the split laser beam; a second delay module configured to receive a second portion of the split laser beam, wherein the first delay module and the second delay module are configured to delay the first portion of the split laser beam and the second portion of the split laser beam such that they do not strike the target simultaneously. In some embodiments, the duty cycle of the laser is doubled.

[0053] In some embodiments, the system further includes a polarization controller configured to tune the first portion of the split laser beam and the second portion of the split laser beam such that they have the same polarization state. In some embodiments, the system further includes a polarization detection module. In some embodiments, the system further includes an optical amplifier configured to increase the power level of the laser beam.

[0054] In some embodiments, the beam steering subsystem is configured to operate at a mechanical scan angle that is 5% to 50% more than the minimum required angle. In some embodiments, the probe surrounds a dichroic mirror that is configured to direct some light from the laser and some light in the visible wavelength to an optical camera. In some embodiments, the optical camera further includes a lens system, wherein the lens system is configured to match the focal plane and the depth of focus of the laser profile. In some embodiments, the aperture size (e.g., diameter) of the OCT lens system in the handheld probe is from about 2 mm to about 15 mm. In some embodiments, the OCT lens system is configured to achieve telecentric scanning of the laser beam. In some embodiments, the aperture of the OCT lens system in the handheld probe is from about 5 mm to about 15 mm. In some embodiments, the OCT lens system is configured to vary the aperture diameter by plus or minus 5 mm to achieve telecentric scanning of the laser beam. In some embodiments, the OCT objective is configured to maintain focus on the target when it is within a range of about 1 mm to about 125 mm from the target. In some embodiments, the optical camera is configured to maintain focus on the target when it is within a range of about 1 mm to about 125 mm from the target. In some embodiments, the OCT objective is configured to maintain focus on the target when it is within a range of about 50 mm to about 125 mm from the target. In some embodiments, the optical camera is configured to maintain focus on the target when it is within a range of about 50 mm to about 125 mm from the target. In some embodiments, the digitizer is a dual-channel signal digitizer.

[0055] In some embodiments, the laser is configured to operate at from about 800 nm to about 2100 nm. In some embodiments, the laser is configured to operate at 1310 nm or 1700 nm. In some embodiments, the plurality of mirrors includes mirrors having dimensions from about 1 mm to about 10 mm.

[0056] Embodiments of a method for generating an image of an object with an optical coherence tomography (OCT) system that can correct for relative motion are provided. The method includes: operating an OCT system that includes a swept-source laser and generates an output laser beam having an optical wavelength range; splitting the laser beam into a sample arm and a reference arm; steering the laser beam onto an object in a scanning pattern; obtaining optical image data from an optical camera positioned within the optical focus of the object, the optical camera configured to be collinear with the imaging path of the laser beam and record optical data of the object; receiving OCT data from the OCT system based on data received from the sample arm and the reference arm; receiving optical imaging data from the optical camera; and correcting the OCT imaging data in one of the following ways to account for relative motion of at least a portion of the OCT system: using the imaging data from the optical camera to correct for the relative motion of a handheld probe that includes the optical camera and the output of the laser beam, and correcting the OCT imaging data based on the determined relative motion of the handheld probe; determining the relative motion from a volume data set by matching successive surface topographies of the object; or acquiring motion data from one or more sensors, the one or more sensors including an inertial measurement unit (IMU), an accelerometer, a gyroscope, a magnetometer, an electromagnetic navigation system, or an optical navigation system.

[0057] In some embodiments, the method further includes stitching together the OCT imaging data and the imaging data of the optical camera to generate combined imaging data that includes a high-resolution image of the object, including volume data below the surface of the object. In some embodiments, the method further includes displaying an overall optical image on a visual interface. In some embodiments, the laser is configured to operate at a sweep duty cycle of from about 35% to about 55%. In some embodiments, the laser is configured to operate at a sweep duty cycle of from about 35% to about 75%. In some embodiments, the laser is configured to operate at a sweep duty cycle of about 75%.

[0058] In some embodiments, operating the OCT system includes operating a handheld probe at a plurality of positions and in a plurality of orientations. In some embodiments, obtaining optical image data from the optical camera includes operating a handheld probe at a plurality of positions and in a plurality of orientations. In some embodiments, operating the OCT system includes operating the OCT system at a plurality of positions and in a plurality of orientations, and wherein obtaining optical image data from the optical camera includes operating the optical camera at a plurality of positions and in a plurality of orientations.

[0059] In some embodiments, the method further includes: converting OCT imaging data into a digital electrical signal; and generating a volume dataset from the digital electrical signal. In some embodiments, the method further includes: correcting scan field distortion in the volume dataset using pre-calibrated reference data to determine the true optical path of each scan beam. In some embodiments, the method further includes: determining features of the target based on the volume dataset; and segmenting the volume dataset based on the features of the target. In some embodiments, the method further includes determining relative motion from the volume dataset by matching consecutive surface topographies of the target.

[0060] In some embodiments, matching consecutive surface topographies includes applying an algorithm including iterative closest point, coherent point drift, normal distribution transform, or a combination thereof. In some embodiments, the method further includes acquiring motion data from one or more sensors, the one or more sensors including an inertial measurement unit (IMU), accelerometer, gyroscope, magnetometer, electromagnetic navigation system, or optical navigation system. In some embodiments, the optical image data includes motion information, the motion information including one or more of the following: orientation, position, linear or angular velocity, linear or angular acceleration of a handheld probe.

[0061] In some embodiments, determining the relative motion of the handheld probe using optical image data of an optical camera includes determining the true path of the laser beam based on the motion information and updating the volume dataset. In some embodiments, determining the relative motion of the handheld probe using optical image data of an optical camera includes applying a block matching algorithm, contour tracking algorithm, cross-correlation, optical flow, or feature-based method.

[0062] In some embodiments, updating the volume dataset includes updating A-scans, B-scans, or C-scans. In some embodiments, updating the volume dataset includes updating A-scans based on determining changes in the position and orientation of the laser beam. In some embodiments, updating the volume dataset includes updating C-scans based on determining changes in the position and orientation of the laser beam.

[0063] In some embodiments, the method further includes correcting motion artifacts in the volume dataset. In some embodiments, the method further includes correcting the refractive index of a target in the volume dataset. In some embodiments, correcting the refractive index in the volume dataset includes addressing the ray shortening effect by calculating the plane normal at each point on the surface of the target, calculating the local slope using surrounding points, and then applying Snell's law to calculate the ray direction. In some embodiments, the target is a tooth, and wherein the refractive index of the tooth is due to enamel or dentin on the tooth. In some embodiments, the target is a tooth, and wherein the refractive index of the tooth is due to the gingiva on the tooth. In some embodiments, correcting the refractive index of enamel will improve the image of the tooth morphology or the image of the tooth below the tooth surface. In some embodiments, correcting the refractive index of enamel includes: detecting the enamel edges in two volume datasets; generating a first array and a second array identifying the 3D coordinates of the enamel edges; applying a transformation matrix to the first array to match the second array by applying deformable registration under constraints; creating a first modified array and a second modified array by removing outliers to generate the first modified array and the second modified array; combining the first modified array and the second modified array together to generate a stitched image. In some embodiments, the method further includes applying a second transformation matrix to the first modified array and the second modified array to improve point matching.

[0064] Embodiments of a dental optical coherence tomography system for performing an intraoral scan of teeth and generating imaging data of the teeth are provided. The system includes: a swept source laser configured to generate an output laser beam having a range of optical wavelengths; a beam splitter configured to receive the laser beam from the swept source laser and split the laser beam into a sample arm and a reference arm; a hand-held probe configured to be inserted into a patient's mouth, the probe surrounding a collimating element that receives the laser beam from the beam splitter, a beam steering subsystem including one or more mirrors configured to change the direction of the laser beam, and a focusing element that receives the laser beam from the beam steering subsystem; an optical camera positioned within the optical focus of the teeth and collinear with the imaging path of the laser beam and recording the optical data of the teeth; a digitizer that receives analog image data and converts the image data into digitized image data; and one or more computer processors configured to implement a method for correcting motion artifacts including one of the following processes: determining the relative motion of the hand-held probe including the optical camera and the output of the laser beam by using the imaging data of the optical camera, and correcting the OCT imaging data based on the determined relative motion of the hand-held probe to correct the OCT imaging data to address the relative motion of at least a portion of the OCT system; determining the relative motion from a volume data set by matching the continuous surface topography of the teeth, and correcting the OCT imaging data based on the determined relative motion of the hand-held probe; or acquiring motion data from one or more sensors including an inertial measurement unit (IMU), an accelerometer, a gyroscope, a magnetometer, an electromagnetic navigation system, or an optical navigation system, and correcting the OCT imaging data based on the determined relative motion of the hand-held probe; or applying a scan pattern that includes steering the beam within a raster sawtooth one-way scan path, a raster scan triangular one-way path, a raster scan two-way scan path, a raster scan sine one-way path, a raster scan sine two-way scan path, a raster scan arbitrary waveform one-way path, a raster scan arbitrary waveform or two-way scan path, a Lissajous scan path, a spiral scan path, a circular scan path, or a radial path, thereby taking multiple frames in an irregular pattern and stitching the multiple frames into a single image, thereby increasing the sampling density.

[0065] In some embodiments, the system corrects for the movement of the handheld probe by providing an A-scan rate of at least 100,000 Hz and providing a synchronization signal, a reference signal, or a trigger signal for each A-scan. In some embodiments, the reference signal includes data showing the scan path of the display laser as a function of time. In some embodiments, the reference signal includes a Mach-Zehnder signal, an interferometer reference signal, a K clock signal, or a combination thereof. In some embodiments, the beam steering subsystem including one or more mirrors includes a fast scan subsystem, which includes a resonant scanner, a galvanometer system, a polygon mirror scanner, a microelectromechanical system, a micro mirror scanner (MEMS), or a combination thereof.

[0066] In some embodiments, the system corrects for the relative movement of the handheld probe by correcting the digitized image data, thereby correcting for the relative movement of the handheld probe when generating the image data of the tooth. In some embodiments, correcting the digitized image data to account for the relative movement of the handheld probe includes applying a block matching algorithm, a contour tracking algorithm, cross-correlation, an optical flow algorithm, a feature-based algorithm, a morphological iterative closest point algorithm, a normal distribution transform algorithm, a coherent point drift algorithm, or a combination thereof.

[0067] In some embodiments, the system further includes a buffer system configured to increase the duty cycle of the laser, the buffer system including: a first delay module configured to receive a first portion of the split laser beam; a second delay module configured to receive a second portion of the split laser beam, wherein the first delay module and the second delay module are configured to delay the first portion of the split laser beam and the second portion of the split laser beam such that they do not strike the tooth simultaneously.

[0068] In some embodiments, the system further includes a polarization controller configured to tune a first portion of the split laser beam and a second portion of the split laser beam such that they have the same polarization state. In some embodiments, the probe further encloses a dichroic mirror configured to direct some light from the laser and some light in the visible wavelength to an optical camera. In some embodiments, the optical camera further includes a lens system, wherein the lens system is configured to match the focal plane and the depth of focus of the laser profile, wherein the aperture of the OCT lens system in the hand-held probe ranges from about 5 mm to about 15 mm, and wherein the OCT lens system is configured to achieve telecentric scanning of the laser beam. In some embodiments, the aperture of the OCT lens system in the hand-held probe is from about 5 mm to about 15 mm, and wherein the OCT lens system is configured to vary the aperture by plus or minus 5 mm to achieve telecentric scanning of the laser beam. In some embodiments, the OCT objective is configured to maintain focus on the tooth when it is in the range of about 1 mm to about 125 mm from the tooth. In some embodiments, the process of correcting motion artifacts includes performing (i), (ii), and (iii) of step (e). In some embodiments, the process of correcting motion artifacts includes using the optical image data of the optical camera and operating the hand-held probe at multiple positions and in multiple orientations simultaneously. In some embodiments, the process of correcting motion artifacts further includes using pre-calibrated reference data to correct the scan field distortion in the volume dataset to determine the true optical path of each scan beam.

[0069] In some embodiments, the system further includes: determining features of the tooth based on the volume dataset; and segmenting the volume dataset based on the features of the tooth. In some embodiments, the system further includes: determining relative motion from the volume dataset by matching the consecutive surface topographies of the tooth. In some embodiments, matching the consecutive surface topographies includes applying an algorithm including iterative closest point, coherent point drift, normal distribution transform, or a combination thereof. In some embodiments, the process of correcting motion artifacts further includes correcting the refractive index of the tooth in the volume dataset. In some embodiments, the process of correcting motion artifacts further includes resolving the ray-shortening effect by calculating the plane normal at the fiducial points on the tooth surface, calculating the local slope using the surrounding points, and then applying Snell's law to calculate the ray direction. In some embodiments, the system further includes correcting the refractive index of the enamel for improving the image of the tooth topography or the image of the tooth below the tooth surface.

[0070] In some embodiments, correcting the refractive index of enamel includes: detecting the enamel edge; generating a first array and a second array that identify the 3D coordinates of the enamel edge; applying a mask to the second array to correct the refractive index of the enamel, thereby modifying the second array to generate a second modified array; applying a transformation matrix to the first array using data from the second modified array to generate a third modified array; refining the third modified array by removing outliers and applying deformable registration under constraints; and combining the first modified array and the third modified array together to generate a stitched image.

[0071] In some embodiments, one or more computer processors are further configured to implement the following method, which includes: steering a laser beam within a raster sawtooth unidirectional scan path, a raster scan triangular unidirectional path, a raster scan bidirectional scan path, a raster scan sinusoidal unidirectional path, a raster scan sinusoidal bidirectional scan path, a raster scan arbitrary waveform unidirectional path, a raster scan arbitrary waveform or bidirectional scan path, a Lissajous scan path, a spiral scan path, a circular scan path, or a radial path. In some embodiments, the method further includes taking a plurality of frames and stitching the plurality of frames into a single image to increase the sampling density. In some embodiments, the beam steering subsystem including one or more mirrors configured to change the direction of the laser beam includes a set of dual x-y resonant scanners.

[0072] In some embodiments, a beam steering subsystem including one or more mirrors configured to change the direction of a laser beam includes a single mirror system containing a single mirror, where the single mirror is actuated by a multi-axis actuator, a microelectromechanical system (MEMS) actuator, a piezoelectric actuator, or a voice coil actuator. In some embodiments, a beam steering subsystem including one or more mirrors configured to change the direction of a laser beam includes a dual mirror system containing two mirrors, where the axes of rotation of the two mirrors are not parallel. In some embodiments, the two mirrors are actuated by a multi-axis actuator, a microelectromechanical system (MEMS) actuator, a piezoelectric actuator, or a voice coil actuator. In the claimed system 102, the two mirrors are actuated by a galvanometer scanner, a resonant scanner, a polygon mirror scanner, a rotary scanner, an acousto-optic scanner, or an electro-optic scanner. In some embodiments, the beam steering subsystem includes a deformable lens having an actuator. In some embodiments, the beam steering subsystem includes a Risley prism-based beam steering system. In some embodiments, a beam steering subsystem including one or more mirrors configured to change the direction of a laser beam includes a microelectromechanical system (MEMS) actuator having a single mirror and a biaxial actuator. In some embodiments, a swept-source laser is configured to generate a small-sized output laser beam, which will reduce the backscattered photon collection efficiency, the OCT signal-to-noise ratio, or both. In some embodiments, the diameter of the single mirror is approximately the diameter of the laser beam, which ranges from approximately 0.5 mm to approximately 6 mm, or from approximately 2 mm to 3 mm. In some embodiments, the method further includes operating the system at a scanning speed of approximately 100 Hz to approximately 5000 Hz, or approximately 500 Hz to approximately 1000 Hz. In some embodiments, the method further includes steering the laser beam relative to two scan axes with two sine waveforms having at least a first frequency and a second frequency. In some embodiments, the method further includes configuring the greatest common divisor (GCD) of the two scan frequencies to be approximately 50 Hz to 3000 Hz, or approximately 100 Hz to 1000 Hz. In some embodiments, the ratio of the first frequency to the second frequency is greater than 0.5, and is approximately 0.66, approximately 0.75, or approximately 0.8. In some embodiments, a beam steering subsystem including one or more mirrors configured to change the direction of a laser beam includes two galvanometer scanners, each galvanometer scanner including a motor attached to one or more mirrors. In some embodiments, the method further includes calibrating the Lissajous scan pattern, and the method further includes: synchronizing the A-scan rate with the Lissajous scan rate; locking the Lissajous scan drive waveform to the A-scan by detecting the timing of the A-scan trigger; removing the jitter in the A-scan trigger; sampling the trigger to generate a jitter-free clock; and generating a scanner drive waveform having the jitter-free clock. In some embodiments, when the light source is swept to a predefined wavelength, the A-scan trigger is a scan start signal or an optical feedback.In some embodiments, jitter removal is performed by a phase-locked loop chip or a microcontroller. In some embodiments, the method includes sampling A number of scans less than the number of Lissajous scans to form a beam spot, wherein the coordinate position of the beam spot is known, and identifying the specific position of the laser beam at each beam spot, thereby calibrating the Lissajous scan pattern. In some embodiments, there is an imaging sensor, wherein the imaging sensor is sensitive to the wavelength of the laser beam, optionally, wherein the wavelength is infrared, and wherein the imaging sensor includes InGaAs or Ge. In some embodiments, the diameter of the imaging sensor is greater than or equal to the field of view of the system. In some embodiments, the imaging sensor is positioned within the field of view of the system and near the focal plane of the system.

[0073] Citation combination

[0074] All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference to the same extent as if each individual publication, patent, and patent application was specifically and individually indicated to be incorporated herein by reference. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The innovative features of the present invention are specifically set forth in the appended claims. The features and advantages of the present invention will be better understood by reference to the following detailed description and the drawings that illustrate exemplary embodiments that utilize the principles of the present invention, in which:

[0076] Figure 1 An exemplary embodiment of an OCT system architecture according to some embodiments is depicted;

[0077] Figure 2A An exemplary embodiment of a swept-source OCT system with a volume scan probe design according to some embodiments is depicted;

[0078] Figures 2B to 2D A swept scan method according to some embodiments is depicted;

[0079] Figure 3 An OCT system according to some embodiments is depicted;

[0080] Figure 4 A buffer system for increasing the duty cycle of a light source according to some embodiments is depicted;

[0081] Figure 5 A scan probe in a sample arm according to some embodiments is depicted;

[0082] Figure 6 A data stream composed of different acquisition modules, adjustment modules, calibration modules, processing modules, and display modules according to some embodiments is depicted;

[0083] Figure 7 Depicts exemplary reconstruction results of an original OCT scan according to some embodiments;

[0084] Figure 8 Depicts the stitching and fusion process according to some embodiments;

[0085] Figure 9A and Figure 9B Depicts the combination of multiple scans according to some embodiments;

[0086] Figure 10 Depicts the stitching of additional volume data to an existing stitched volume data set according to some embodiments;

[0087] Figure 11 Depicts an intraoral scanner with a full-field OCT configuration according to some embodiments; and

[0088] Figures 12 to 18 Depicts a scanning probe according to some embodiments.

[0089] FIG. 19 is a schematic block diagram of an optical coherence tomography system according to the prior art.

[0090] Figure 20 Shows a line segment on the surface of a measured object according to the prior art.

[0091] Figure 21 Shows an exemplary combination of multiple A-scans generating a two-dimensional reflectivity profile along a B-scan according to the prior art.

[0092] Figure 22 Shows an exemplary combination of multiple A / B-scans generating a three-dimensional reflectivity profile along a C-scan using raster scanning according to the prior art.

[0093] Figure 23 Shows a conventional grating according to the prior art.

[0094] Figure 24 Shows the problems (motion blur) exhibited by a conventional raster scan according to the prior art due to in-frame movement of the scan bar.

[0095] Figure 25 Shows a hypothetical conventional OCT system densely scanning a line segment (composed of individual sampling points (pixels)) and a hypothetical sparse scan line segment of an embodiment of the present invention.

[0096] Figure 26 Shows an exemplary Lissajous figure used in an embodiment of the present invention.

[0097] Figure 27A schematic block diagram of a tomographic scanning system according to an embodiment of the present invention is shown. The illustration in FIG. 9 shows an aspect of using the tomographic scanning system in FIG. 9.

[0098] Figure 28 Other exemplary Lissajous figures that can be used in embodiments of the present invention are shown.

[0099] Figure 29 The Lissajous figure in FIG. 9 is mainly shown in dashed lines to show the concept of the line segments related to the embodiments of the present invention.

[0100] Figure 30 Multiple partially overlapping scan patterns according to an embodiment of the present invention are shown.

[0101] Figure 31 The use of an embodiment of the present invention is shown, and specifically shown is a scanning rod that is translated along a path close to the anatomical structure to be measured in space. Detailed Description

[0102] Optical Coherence Tomography (OCT) systems employ an imaging technique that directs a probe beam onto an object and detects backscattering. They typically use a point scanning mechanism that forms a single beam of sampling light in open air. When the sampling beam is incident on the object surface, the beam can also propagate into the object and form an optical path. OCT captures information from the optical path. The point scanning mechanism then moves the beam to different positions on the object surface and forms different optical paths. Points on the object surface are sampled in sequence, after which some information about the object can be recorded.

[0103] Common requirements for the scan path are high geometric accuracy, high reproducibility, high duty cycle, a uniformly distributed scan pattern, and high speed. High geometric accuracy is required to recover the true structure. Sometimes it is also highly desirable to calibrate the scan path to improve accuracy. High reproducibility is required to accurately reconstruct the geometry of the object when scanning repeatedly. A high duty cycle is required to scan as many sampling points as possible. A roughly uniform scan pattern is required to reduce sampling errors for complex structures. High speed is required or even essential when the object moves relative to the scanner.

[0104] The present disclosure describes some exemplary embodiments of an OCT scanning system that has an ultrafast scan speed, high reproducibility, and high geometric accuracy, while taking into account the scan point distribution and duty cycle.

[0105] There are technical limitations in obtaining high-quality OCT images. First, OCT typically has a short working distance on the order of millimeters and a tissue penetration depth of 1 millimeter to 10 millimeters. Different from X-rays or CTs where the imager can be placed outside the oral cavity, the OCT scanner usually has an intraoral part close to the teeth during scanning. In some embodiments, a handheld design is required, which places restrictions on the weight and size of the scanner. Second, OCT is typically used in a laser scanning configuration. Although existing OCTs may be able to provide a high A-scan rate, motion of the operator or patient may introduce significant image distortion. Careful decisions should be made to minimize and correct such motion artifacts. Third, due to the limited size and sampling density of the intraoral part, the field of view of an OCT scan is typically about 10×10×10 cubic millimeters. Variations are applicable to different system designs. Throughout the specification, the volume from which OCT obtains data is referred to as the OCT scan volume. To scan an entire tooth or multiple teeth, multiple scan volumes may need to be acquired and the volume data stitched (e.g., registered and fused) together. The system hardware and software need to be carefully designed to facilitate such stitching algorithms.

[0106] Figure 1 An exemplary embodiment of the OCT system architecture is depicted. In some embodiments, the system includes a handheld scanning probe 11 and a system engine 12. In some embodiments, the handheld scanning probe surrounds the sample arm of the OCT system. In some embodiments, a probe beam profile is formed to meet specific characteristics and the beam is rapidly modified in a predefined pattern such that the acquired image data can be used for reconstruction. In some embodiments, photons backscattered from the tissue surface carry information for reconstructing the surface topography. Photons backscattered from internal tissue can carry information for visualizing the internal tissue structure or function. The system engine may also include a reference arm, an interferometer, a light source, a detection subsystem, and their respective control modules. The image data can then be digitized and transmitted to a computer for further analysis and visualization.

[0107] The present disclosure can address the above needs by providing an OCT scanning system with an ultrafast scanning speed, thereby reducing motion artifacts. Figure 2ADepicts an exemplary embodiment of a swept - source OCT system with a volumetric scanning probe design according to some embodiments. Swept - source OCT systems typically have a high volumetric scanning rate, which can reduce motion artifacts. Some embodiments of swept - source OCT systems involve a long coherence length, which can allow for minimal signal - to - noise ratio degradation and a long imaging range. In some embodiments, it allows for imaging the entire anatomy of teeth with a height variation greater than 10 mm. In some embodiments, the swept - source 21 emits light that sweeps over a specific wavelength range into a beam splitter 29. In some embodiments, the light is then split into a sample arm 22 and a reference arm 2001. In some embodiments, the sample arm 22 includes a collimating element 25, two mirrors 23 and 24 with non - parallel axes of rotation, which steer the collimated beam to follow a raster scan pattern or any other suitable scan pattern. In some embodiments, the beam is then refocused by a focusing element 26 to improve the lateral resolution of the OCT imaging. The mirror 28 can be mounted to deflect the scanning beam 27 by the required angle. In some embodiments, the backscattered light from the subject travels back along the incident optical path. In some embodiments, the backscattered light then interferes with the light returning from the reference arm 2001 at the coupler 29. The interference signal is then acquired by the detection module 200 for further analysis.

[0108] Figures 2B to 2D Depicts embodiments of a swept - scan method and explains some terms used in an OCT scanning system. The beam 201 can represent the position and orientation of the OCT scanning beam at time t0 and is incident on the tissue surface 203. As the light beam propagates along a path in the tissue, it can be partially backscattered, travel back, and be detected by the OCT system. This path can be referred to as an A - scan. Each A - scan can correspond to one laser sweep along the swept wavelength. The direction of the A - scan is typically referred to as the axial direction or Z - direction. The beam can then move to the next position 202 at time t0+Δt to capture the next A - scan. The scanning can continue, thus capturing a set of A - scans that represent a tissue cross - section 205, and this scan can be referred to as a B - scan. The scanning direction of the B - scan is transverse to the A - scan and can be referred to as the fast axis. In three - dimensional space, the remaining direction perpendicular to the axial direction and the fast axis can be referred to as the slow axis. Conventional OCT scanning techniques can be further referred to in FIGS. 19 to Figure 24 described. In the raster - scan method, as Figure 2D shown, the beam can follow a zig - zag pattern such that each B - scan is offset with respect to the previous B - scan. A set of B - scans forms volumetric data, which is referred to as a C - scan 206 or volumetric scan.

[0109] In some embodiments, Figure 2ADepicts that the beam profile 27 is a Gaussian profile. In some embodiments, the focal plane is the transverse plane (relative to the light propagation direction) where the beam waist is located. The 1 / e 2 diameter (also known as the spot size) can be in the range of 1 μm to 50 μm. When performing a uniformly spaced transverse scan, the distance between two adjacent sampling points can be in the range of 1 μm to 200 μm.

[0110] In some embodiments, the OCT system architecture can be configured to more effectively utilize light to improve the signal-to-noise ratio. In some other embodiments, the OCT system can be configured to utilize the polarization state of light to extract tissue features and improve image quality.

[0111] Figure 3 Depicts an exemplary embodiment of an OCT system for polarization state extraction. Since enamel is a birefringent material, a polarization-sensitive OCT system can be particularly useful when imaging enamel. Polarization-sensitive OCT can detect birefringence or can be used to remove birefringence artifacts in an image to easily detect other features in enamel. In some embodiments, the swept-source laser 301 is connected to the coupler 302. In some embodiments, to improve power efficiency, more power can be delivered to the sample arm to increase the signal strength while keeping the emission power below the accessible emission limit (AEL) safety limit and retaining sufficient optical power in the reference arm to fully utilize the full dynamic range of the detection module in the system. The AEL is the maximum accessible level of laser radiation allowed within a specific laser class. In some embodiments, the splitting ratio of the coupler ranges from 99% / 1% to 70% / 30%. In some embodiments, the splitting ratio of the coupler ranges from 95% / 5% to 90% / 10%. Polarization controllers 305 and 306 can be added to the sample arm and the reference arm respectively to adjust the polarization state of light for extracting polarization-sensitive information. Additional polarization controllers (not shown) can be added before the polarization couplers 309 and 310. These polarization couplers can split the light from the sample arm and the reference arm into orthogonal states and then independently interfere at the 50 / 50 couplers 311 and 312 and are detected by independent detectors 313 and 314. In some embodiments, a dual-signal channel digitizer 315 is used in the computer 316 to detect the signals. The dual-signal channel digitizer can reduce the space requirement of the computer. Repeated sampling of the OCT signal may require a reference clock from the laser, which shows the relationship of the swept wavelength with time. This reference clock is also fed to the digitizer 315.

[0112] In some embodiments, the swept source is configured to specifically meet the requirements of dental imaging. The accuracy of the OCT system can be sufficient to identify features of interest in teeth, such as the morphology of the teeth or dental caries disease. Therefore, the 15 dB bandwidth of the light source can be greater than 50 nm, such that the axial full width at half maximum (FWHM) of the OCT imaging can be small enough.

[0113] In some embodiments, the center wavelength of the swept source is in the near-infrared region, more specifically from 800 nm to 2.5 μm, such as 840 nm, 1060 nm, 1310 nm, 1700 nm, 2100 nm, etc. In some embodiments, the center wavelength is approximately 1310 nm, such that the system can generally penetrate deeper into biological tissue compared to other shorter wavelengths. In some embodiments, a swept source around 1310 nm is more readily available. In some embodiments, a center wavelength around 1700 nm is used to reduce the water absorption rate.

[0114] In some embodiments, the swept source laser includes a wavelength sweep duty cycle equal to or less than 50%. In other words, the swept source laser can have an emission cycle, and the duration during which the swept source laser emits useful light can be equal to or less than half of that cycle. In these lasers, a buffer system can be used to increase the duty cycle of the light source system and effectively increase the scanning rate of the system, as depicted in Figure 4 The graph of power versus time depicts the light emission timing and power of each component.

[0115] In some embodiments, the light from the swept source 401 can be split into two parts by a 50 / 50 coupler 402 and enter the buffer system. In some embodiments, the first part of the split light travels through a first delay module 403. In some embodiments, the second part of the split light travels through a second delay module 404. In some embodiments, the difference between the first delay and the second delay is configured such that when the light from modules 403 and 404 recombines in a second 50 / 50 coupler 405, the light with the first delay and the light with the second delay do not overlap in time. In this configuration, the combined duty cycle of the light source system is effectively doubled compared to the swept source itself. In some embodiments, the optical delay can be introduced by a long waveguide (such as an optical fiber coil).

[0116] In some embodiments, an optical amplifier may be used in combination with a beam splitting configuration such that a first portion and a second portion of the split light can be tuned to a desired power level. In some embodiments, a polarization controller 408 and a polarization detection module 409 are used in combination with the beam splitting configuration such that a first delayed portion and a second delayed portion of the split light can be tuned to the same polarization state. This may be particularly important for dental imaging because enamel is a birefringent material and is sensitive to the polarization state of light. In some embodiments, the optics of the first delay module 403 and the second delay module 404 affect the output polarization state of the light. In some cases, when the two portions of the light have different polarization states without any compensation, the system emits light of alternating polarization states every other A-scan, which introduces different polarization artifacts every other A-scan in the image and produces an image texture with artifacts.

[0117] Figure 5 An exemplary embodiment of a scanning probe in a sample arm is depicted. The sample arm light can be calibrated and directed into a beam steering module 501. The beam steering module 501 may have two mirrors 502 that serve as beam reflecting elements. In some embodiments, the mirrors 502 are mounted on oscillating components (such as two galvanometer motors), which form a galvanometric mirror subsystem. The galvanometric mirror subsystem can be used for its defined motion feedback and control as well as scanning linearity, which may be required for simplified data reconstruction in subsequent processing stages. The oscillation axes of the two mirrors can be perpendicular to facilitate position control such that one mirror controls the sweep frequency of the beam in one direction while the other mirror controls the sweep frequency in a direction orthogonal thereto. In some embodiments, the galvanometric mirror subsystem is configured to scan not only as fast as possible but also with minimal scan field distortion and high reproducibility. Thus, the subsystem can be configured for one-way scanning and only use the linear scanning portion. In some embodiments, the size of the collimated beam incident on the galvanometric mirror determines the size of the mirror. In some embodiments, the size of the mirror determines the linear scanning frequency range, which in turn determines the maximum mechanical scanning speed. In some embodiments, the size of the collimated beam can range from 1 mm to 3 mm to obtain a sufficient lateral spot size in tissue scanning, which is beneficial for tissue feature identification and measurement accuracy. In some embodiments, the size of the galvanometric mirror ranges from 3 mm to 7 mm. In some embodiments, the scanning frequency range is from 100 Hz to 400 Hz such that the galvanometer motors are still in the linear scanning range but as fast as allowed by the subsystem.

[0118] In some embodiments, the beam steered by the galvanometric mirror system is then refocused by the lens system 504, reflected by the mirror 505, and exits through the imaging window 506. A transparent window 508 may be installed to prevent contamination of the lens system 504. In some exemplary embodiments, the OCT focal plane may be configured to be 1 mm to 5 mm from the outer surface of the cannula 507. This may allow the cannula 507 to be kept relatively close to the target tooth without overstretching the subject's cheek or causing the subject to open their mouth wide. In some embodiments, for the comfort of the subject, it may be desirable to maintain a low-profile cannula 507. To achieve a low cannula profile, lenses with a small aperture and a long focal length may be used. In some embodiments, the aperture of the lens 504 is approximately 5 mm to 15 mm. In some embodiments, the focal length of the refocusing lens system is approximately 50 mm to 125 mm. In some embodiments, the galvanometric mirror 502 is positioned at a distance from the refocusing lens system 504 within the range of + / - 5 mm of the focal length to achieve or approximate telecentric scanning of the OCT beam. The optical elements in the lens system 504 may be any combination of a single lens, an achromatic lens, a lens assembly, a diffractive lens, a metalens, or other suitable optical elements. The optical elements may be made of any suitable material.

[0119] In some embodiments, a video camera visualizes the tissue scanned by the OCT. OCT images require 3D reconstruction, and real-time OCT visualization may require high computational power. Therefore, real-time visualization of OCT data may not be cost-effective, and video camera visualization may be used as a supplement or alternative to real-time OCT visualization. In some embodiments, the system utilizes video camera motion tracking technology to correct motion distortion. Figure 5 A video camera configuration collinear with the OCT imaging path is depicted, which is achieved by the dichroic mirror 504 and the video camera 503. The dichroic mirror 504 is configured to transmit OCT light and reflect visible light. In some embodiments, the visible light is reflected or scattered by the tooth and redirected to the camera 503. In some embodiments, the lens system in the camera 503 is configured to match the focal plane and the depth of focus of the OCT beam profile. For example, when the OCT focal plane is 2 mm from the outer profile of the cannula 507 and the depth of field is 2 mm, the tooth surface of interest may be arranged to be approximately 1 mm from the cannula 507 such that the depth of field is within the tooth. In this example, the focal plane of the video camera may be set to be 1 mm to 2 mm from the cannula 507 such that when the video camera image is in focus, the OCT imaging is close to its ideal range.

[0120] Scanner actuator

[0121] In one embodiment, the control of the optical scanning beam is achieved by using at least one mirror. The scanning path can be controlled by changing the orientation of the mirror and thus changing the reflection direction of the beam propagation. One mirror or multiple mirrors can be electric and can be programmed to change the orientation of the mirror surface in a defined pattern.

[0122] In some embodiments, there can be a mirror system where the mirror can be mounted on a biaxial or multi-axis actuator, such as a microelectromechanical system (MEMS) actuator, a piezoelectric actuator, a voice coil actuator, etc.

[0123] In some other embodiments, there can also be two or more mirror systems. The axes of rotation of these mirrors do not have to be perpendicular. The type of actuator for each mirror can be the same or different. The actuator can be one of the above-mentioned actuators. In addition, the actuator can also be a galvanometer scanner, a resonant scanner, a polygon mirror scanner, a rotary scanner, an acousto-optic scanner, an electro-optic scanner, etc.

[0124] In some other embodiments, the control of the optical scanning beam is achieved by using a deformable lens with an actuator. In still some other embodiments, the control of the optical scanning beam is achieved by a Risley prism-based beam steering system.

[0125] Scanning pattern

[0126] In some embodiments, when the scanning pattern is a raster scan, the first galvanometer mirror scans along the fast axis, while the second galvanometer mirror scans along the slow axis. The OCT beam guided into the galvanometer mirror subsystem can be incident on the first mirror on the fast axis and then be reflected to the second mirror on the slow axis. In some embodiments, this allows reducing the effective area of the mirror, thus reducing the mass oscillating along the fast axis, reducing mirror deformation, and increasing the scanning speed of the fast axis.

[0127] Ultra-fast scanning path and configuration

[0128] Ultra-fast scanning speed is highly desirable for reducing motion blur. This can be achieved by a combination of selecting high-speed actuators and adopting linear or non-linear scanning trajectories, limiting actuator overshoot or mirror deformation, selecting sampling density and field of view (FOV), etc.

[0129] In some embodiments, ultra-fast speed is achieved by having a fast scanning axis and a high scanning frequency. The actuator for the fast-axis mirror can be driven using a high-frequency sine waveform. As long as the actuator or system overshoot is acceptable, other harmonics can be added to the drive waveform to improve linearity. In one embodiment, a polygon mirror scanner is used in the fast axis.

[0130] In some embodiments, the other scanning axis is a slower scanning actuator and mirror. The combination of the fast scanning axis and the slow scanning axis produces a raster scan trajectory, which can be unidirectional or bidirectional. In some other embodiments, the other scanning axis can also be scanned at a high frequency but at a frequency different from that of the first axis, which will form a Lissajous scan pattern. In yet another embodiment, the two axes can have the same high frequency but different scanning angles, which will form a helical scan pattern.

[0131] In some embodiments, when a high OCT volume rate is required, the mirror along the fast axis may need to be scanned up to 10,000 Hz. In some embodiments, the scanning probe design may consider the laser A-scan rate, the incident parallel beam size, and the scanning field of view. In some embodiments, the laser A-scan rate determines the maximum scanning rate that the system can achieve. In some embodiments, the beam size determines the scanning lateral resolution and the depth of field. In some embodiments, the scanning field of view determines the area that the scanning volume can cover. For example, in a Gaussian beam configuration, the projection of the mirror along the beam direction may be approximately 1 / e of the beam diameter. 2 In some embodiments, the projection of the mirror along the beam direction may be approximately 1% of the beam diameter. In some embodiments, the projection of the mirror along the beam direction may be approximately 0.1% of the beam diameter. In some embodiments, the mechanical beam steering angle should match the scanning field of view so as to scan an additional angle for easy optical alignment. In some embodiments, scanning the additional angle causes the laser not to scan the field of view and reduces the duty cycle. Therefore, the mechanical scanning angle can scan an additional 5% to 50% of the minimum required angle to achieve the field of view.

[0132] In some embodiments, the fast scanning mirror is mounted on a resonant scanner to achieve higher speeds. In some embodiments, the resonant scanner is used for the sinusoidal characteristic of its scanning angle. In some embodiments, the resonant scanner operates at a fixed frequency. In some embodiments, the frequency and phase of the scanning sine wave can be used to calibrate the scanning non-linearity. In some embodiments, the scanning frequency can be configured from 1,500 Hz to 10,000 Hz. For example, when using a 200,000-Hz A-scan swept source and capturing 20 volumes per second, each volume consists of 10,000 A-scans. In some embodiments, a scanning pattern can be configured such that each B-scan includes 100 A-scans and each volume includes 100 B-scans. In this exemplary embodiment, the resonant scan can be configured to 2,000 Hz.

[0133] In some other embodiments, the fast steering mirror includes a multi-faceted rotating mirror to achieve higher speeds. In some embodiments, a multi-faceted mirror scanner is used for the linearity of its scanning angle. In some embodiments, the rotation speed and the scan start signal of the multi-faceted scanner are used to calibrate the scan field. In some embodiments, ball bearings are used to allow the scanner to be oriented in any direction.

[0134] In some embodiments, the multi-faceted mirror scanner with ball bearings has a maximum revolutions per minute (RPM). In some embodiments, the appropriate number of facets and RPM of the multi-faceted mirror surface should be matched with the rest of the scanning optics to maximize the scan duty cycle, because in some embodiments, the higher the scan duty cycle, the faster the system can acquire data, thus generating fewer motion artifacts. In some embodiments, if the scanning optics requires an optical scan angle of + / -5° with respect to the fast axis, the mechanical scan angle required for the multi-faceted mirror is halved, i.e., 5° in total for each facet. In some embodiments, additional mechanical scan angles are used for tolerance stacking and introducing the scan start signal. In some embodiments, the angle of each facet of the multi-faceted mirror can be 6°, which results in a multi-faceted mirror having 60 facets (360° / 6°). In some embodiments, to achieve a fast axis scan rate of, for example, 2000 Hz, the multi-faceted mirror can be configured to rotate at 2000 RPM. In some embodiments, the mechanical scan angle is 101% to 200% of the required optical scan angle.

[0135] In some embodiments, a mirror mounted on a microelectromechanical system (MEMS) can be used to replace one or two mirrors on the fast axis. In some embodiments, a small MEMS mirror can be configured to scan on two axes. By reducing the size of the mirror and limiting the scan angle, the MEMS mirror can be configured to scan at a frequency of thousands of kilohertz.

[0136] In some embodiments, when the OCT system uses a swept-source laser as the light source, the sampling frequency h can be determined by multiple factors, such as, for example, the A-scan rate, the center wavelength, the swept bandwidth, the laser sweep duty cycle, the beam steering duty cycle, and the imaging depth, etc. In some embodiments, the reflected optical signal is digitized by an A / D digitizer. In some embodiments, as long as the sampling rate is within the capacity of the digitizer, the A-scan rate is as high as possible to minimize motion artifacts in the C-scan. In some embodiments, the scanning speed of a high-end swept-source laser can exceed 1,000,000 A-scans per second.

[0137] In some embodiments, the system configuration includes a field of view of at least approximately 6 mm × 6 mm using raster scanning and 10 C-scans per second. In some embodiments, a field of view of 8 mm × 8 mm using raster scanning and 25 C-scans per second is provided. In some embodiments, a higher C-scan rate is required because each C-scan will have fewer motion artifacts. In some embodiments, the lateral spacing between A-scans is optimized by considering the average motion speed of the scanning probe, or the lateral sampling resolution is optimized. In some embodiments, if the average motion speed is high, this means fewer C-scans for data capture per tooth. In some embodiments, a higher lateral sampling resolution will ensure that the system has sufficient data to reconstruct the tooth. In one example, assuming the sampling resolution of the system is 100 μm, with a field of view of 8 mm × 8 mm and 25 C-scans per second, each C-scan has a scanning pattern of at least 80 points × 80 points (8 mm / 100 μm). In some embodiments, square scanning is used for isotropy on the fast and slow axes.

[0138] In some embodiments, the raster scanning system uses various techniques (e.g., galvanometric scanning mirrors) to scan on the slow axis with a duty cycle of approximately 100%. On the other hand, due to mirror retrace / reset (such as in galvanometric, resonant, or MEMS scanners), or due to the need for additional time to identify the start of the scan (such as in polygon scanners), the duty cycle of the fast axis may be less than 100%. Therefore, the actual A-scans for each C-scan can be estimated as 80 * 80 / (fast axis duty cycle). In some embodiments, if a 75% duty cycle is used, each C-scan requires at least 8534 A-scans, and the A-laser sweep frequency is at least 213,333 Hz. Thus, the sweep frequency of the sweep source can be at least 100,000 Hz, 200,000 Hz, or higher.

[0139] Scanning pattern

[0140] In other embodiments, various beam steering / scanning paths can be applied, including but not limited to raster scanning with a zigzag unidirectional scanning path, raster scanning with a triangular unidirectional or bidirectional scanning path, raster scanning with a sinusoidal unidirectional or bidirectional scanning path, raster scanning with a unidirectional or bidirectional scanning path of any waveform, Lissajous paths, spiral paths, circular paths, radial paths, etc. In some cases, applying an alternative scanning pattern as described herein can reduce motion artifacts caused by movement of the subject during scanning. In a standard raster scan, the distance between the first pixels in each row (e.g., at 0,1; 0,2; 0,3 in the x-y coordinate plane) and the distance between the first and last pixels scanned take the longest time between scans and may create significant motion artifacts between the pixels rendered by these scans. When scanning in an alternative pattern as described herein, the maximum time delay no longer occurs between pixels in adjacent rows, but rather between pixels at irregular intervals around the target being scanned.

[0141] For example, in a standard raster scan, a laser sweeps (covers) a 2D area and uses two significantly different scan frequencies: a fast scan frequency along what is typically a horizontal line and a slow scan frequency along the y-axis. Typically, the second scan remains constant during the sweep of the first scan. For example, the first fast scan is performed in the horizontal (x) direction while there is no movement in the second vertical (y) direction. This results in a high sampling rate in the first direction with little motion artifact. However, across the entire frame, depending on the scan rate in the second direction, there may be a significant delay between the first and last rows, which can lead to motion artifacts across the entire frame. The time delay increases gradually along the slow scan axis, which can cause a drift in motion error in the diagonal direction. This gradual drift can pose a problem for stitching scanned images of multiple volumes, and if not corrected, this drift may accumulate in subsequent scans.

[0142] A Lissajous scan pattern or other alternative scan patterns described herein can effectively redistribute the scan points in the scan field, thus interweaving scan points with large and small motion artifacts. The sample can be imaged across the entire frame at a rate that is the greatest common divisor of two scan frequencies. Although the sampling of a single "image" is sparser across the entire frame compared to raster scan methods, motion artifacts can be significantly reduced or eliminated. Denser sampling is achieved by taking multiple frames and registering (or stitching) them together to achieve the desired coverage and sampling density. Motion artifacts can be further reduced by applying a smoothing filter or performing multiple scans with a small spatial displacement. The Lissajous scan or other alternative scan patterns described herein can be implemented using a multifaceted mirror, a galvanometric scanner, a resonant scanner, or a MEMS scanner, or other scanners and reflectors known in the art, and are not limited to the structures described herein. Alternative configurations of the OCT system can be used to support these alternative scan patterns. In some embodiments, the system can include a dual x-y resonant scanner.

[0143] MEMS Lissajous pattern

[0144] In one exemplary embodiment, the optical scanning system is a MEMS scanner with a single mirror and a biaxial actuator. The single mirror system has the potential to greatly reduce the required space. Generally, the larger the mirror size, the slower the scan speed. Therefore, a small mirror is preferred for ultrafast scanning. On the other hand, the mirror size needs to be at least similar to or larger than the optical scan beam size, thus setting an upper limit on the beam size. A smaller beam size generally reduces the backscattered photon collection efficiency of OCT, which will reduce the OCT signal-to-noise ratio. An appropriate mirror size should be determined to balance the trade-off between scan speed and signal-to-noise ratio. The mirror size can be determined empirically.

[0145] In one embodiment, the mirror size can be from 0.5 mm to 6 mm, preferably 2 mm to 3 mm. The scan speed can be from 100 Hz to 5,000 Hz, preferably 500 Hz to 1,000 Hz.

[0146] In one embodiment, the scan pattern can be sinusoidal along the fast scan axis and linear along the slow scan axis, which will form a raster scan pattern. In another embodiment, the two scan axes can be two sine waveforms with different frequencies. Highly preferably, the greatest common divisor (GCD) of the two scan frequencies is configured to be approximately 50 Hz to 3000 Hz. The GCD is the frequency of the scanner that performs the Lissajous pattern. In one embodiment, the GCD can be 100 Hz to 1,000 Hz. The ratio between the slower scan frequency and the faster scan frequency is preferably greater than 1 / 2. A higher ratio allows the Lissajous pattern to scan the FOV more uniformly, thus reducing the sampling error. In one embodiment, the ratio between the slower scan frequency and the faster scan frequency can be 2 / 3, 3 / 4, 4 / 5, etc.

[0147] Galvo Lissajous pattern

[0148] In another exemplary embodiment, the optical scanning system consists of two galvo scanners, each with an electrically driven mirror. Although galvo scanners are larger in size, for the same size of mirror, they can generally achieve higher scan speeds. The above mirror size and scan frequency specifications also apply.

[0149] Ultra-fast Lissajous scan calibration

[0150] When high calibration accuracy is required, the calibration of the Lissajous pattern can be challenging. It is known that the precise positioning of each A-scan can be affected by different error sources. There may be synchronization errors between the motor control signal and the OCT A-scan. The scan mirror may deform during scanning. There will also be scan field distortion and aberration. In a dense and uniformly distributed scan pattern, the calibration is relatively simple. A common calibration method is to scan a model with a known geometry. Since the scan pattern is dense and uniform, the marks on the model can be directly identified and used for calibration. In the case of the Lissajous pattern on the other hand, the scan trajectory is less well-defined. The FOV scan is not uniform. In an ultra-fast Lissajous pattern, the scan field is sparsely scanned to increase the scan speed. It may be difficult to register the scan data to spatial coordinates to reconstruct the object. Calibration can still be performed if the Lissajous pattern is repeatable and the spatial position of the scan beam can be identified when capturing the A-scan.

[0151] To make the Lissajous pattern repeatable, in an exemplary embodiment, the Lissajous scan should be synchronized with the OCT A scan rate. The OCT A scan rate is typically much higher than the Lissajous scan. The system can be configured such that the Lissajous scan drive waveform is phase-locked to the OCT A scan. In this way, the sampling positions in each Lissajous period are highly repeatable. In an exemplary embodiment, phase locking can be achieved by detecting the exact timing of the A scan trigger, removing the jitter in the trigger, and upsampling the trigger to generate a jitter-free clock, which is then used to generate the scanner drive waveform. In a swept-source OCT system, the A scan trigger can be the scan start signal or the optical feedback when the light source scans to a specific wavelength. Jitter removal and upsampling can be accomplished by a phase-locked loop chip, a microcontroller, etc.

[0152] To identify the spatial position of the A scan, in one embodiment, the calibration for the ultrafast Lissajous scan pattern involves using OCT imaging and an imaging sensor. The OCT imaging system can be a system with an ultrafast Lissajous scanner. The imaging sensor can be those sensors sensitive to the wavelength of the OCT system. Common OCT systems use infrared wavelengths, so the imaging sensor can be based on Si, InGaAs, or Ge. The sensor can be a linear array or preferably a area array. The sensor preferably has a global shutter and a short exposure time.

[0153] In an exemplary embodiment, the image sensor is selected such that the image size of the sensor is equal to or greater than the FOV of the OCT system. The image sensor is positioned in the FOV such that the OCT FOV falls within the sensor area. The image sensor is preferably positioned near the focal plane of the OCT scanning optics such that the spot size of the scanning beam in the sensor is small. The small spot size improves the detection accuracy of the beam position.

[0154] The exposure of the image sensor should also be synchronized with the OCT A scan rate. The purpose is to trigger the sensor exposure at a specified position on the Lissajous trajectory. Due to the short exposure time, this triggering mechanism enables the sensor to capture only a few OCT A scans. These few OCT A scans only account for a part of the Lissajous period, which appears as a dot curve with a starting point and an ending point in the imaging sensor. The coordinates of these beam spots can be used to identify the spatial position of the scanning beam of each A scan. By triggering different segments of the Lissajous pattern, a complete calibration of the Lissajous pattern can be performed.

[0155] In some embodiments, the imaging depth of the OCT is at least 5 mm, 10 mm, or higher.

[0156] In some embodiments, the -15dB wavelength bandwidth of the swept source is approximately 30nm to 200nm. In some embodiments, the smaller the bandwidth, the lower the maximum RF frequency. However, a smaller wavelength bandwidth may result in a larger point spread function, which will reduce the OCT axial resolution ability. Therefore, the selected bandwidth may need to be a trade-off between axial resolution ability and RF signal bandwidth.

[0157] In some embodiments, the duty cycle of the swept source without using a buffer system is between 40% and 70%. In some embodiments, the higher the scan duty cycle, the lower the maximum RF frequency. In some embodiments, as Figure 4 depicted in, using a buffer system further reduces the maximum RF frequency.

[0158] In some embodiments, after configuring the A-scan rate, center wavelength, scan bandwidth, laser sweep duty cycle, beam steering duty cycle, and imaging depth, the RF signal bandwidth can be estimated. In some embodiments, the RF signal bandwidth is 500MHz or higher. In some embodiments, a fast digitizer with a bandwidth of at least 500MHz and a sampling rate of 1GS / s may be required. In some embodiments, due to the accuracy requirements of the system, synchronization between the laser, beam steering system, and digitizer is crucial. In some embodiments, the digitizer includes additional input ports and receives start signals for A-scans, B-scans, and C-scans for subsequent 3D reconstruction.

[0159] Figure 6 An example of the data stream is depicted, which consists of different acquisition modules, conditioning modules, calibration modules, processing modules, and display modules. The modules can be hardware or software subsystems. In some embodiments, the OCT system includes a detection module 601 that is used to convert the detected photons into electrical signals for further processing. This conversion can generate an electrical analog signal, which can then be digitized in the digitization module 602. The digitization module 602 can receive many reference, trigger, and synchronization signals from the scanning system to achieve high image quality and find the spatial distribution of the scan pattern. In some embodiments, the swept source OCT system can receive a time reference for wavelength sweeping to correct wavelength sweep non-linearity. The OCT system herein is capable of performing high-precision dental scans and taking into account the steering pattern of the scan beam for accurate structural reconstruction. In some cases, the digitization module 602 can receive scan start and end synchronization signals for each beam steering axis. In some embodiments, the output of this module is volume raw data.

[0160] In most scanning systems, scan field distortion cannot be ignored. In some embodiments, the scan field correction module 603 uses pre-calibrated reference data to correct scan field distortion. In some embodiments, the scan field correction module calculates the way the scan beam steers during operation. In some embodiments, a motion artifact correction module 605 is provided. In some embodiments, the scan probe remains stationary relative to the teeth. In this instance, the known beam scan pattern is equivalent to the spatial coordinates of the known tissue being scanned. Relative movement between the scan probe and the teeth is inevitable. In some embodiments, determining the spatial coordinates of the tissue being scanned takes into account the movement, which is the purpose of the motion artifact correction module 605.

[0161] The scan beam may also be affected by interactions with other objects in the scan field, which is determined by the refractive index. In some embodiments, the refractive index correction module 606 is used to determine the propagation path of the beam in tissue or other materials. The known true propagation path may be crucial for reconstructing the true tissue anatomy. In some embodiments, the segmentation module 604 is used for the volumetric raw data from the digitization module 602, for the motion correction data 605, and / or for the reconstruction data from the refractive index correction module 606. Figure 6 An example of the way of processing data according to some embodiments is depicted. In some embodiments, the order or data flow of the modules 603, 604, 605, and 606 can be configured in different ways by those skilled in the art as long as the distortion is corrected and the tissue features are segmented.

[0162] In some embodiments, the scanning system continuously acquires a 3D data volume. In some embodiments, the volume is captured using the scan probe at different positions and orientations. In some embodiments, a registration module 607 is provided within the data processing pipeline to determine the spatial relationships of the 3D data volumes and fuse the data into a meta 3D dataset of one or more teeth. In some embodiments, the display module 608 is used to visualize the information available for decision-making.

[0163] In some embodiments, the motion correction module 605 obtains motion information of the scanning probe and / or the patient to correct image distortion introduced by such motion during imaging. In some embodiments, such motion is referred to as bulk motion. In some embodiments, since the OCT system uses a scanning technique, there is a time delay from one scanning point to the next, although the delay may be small. In some embodiments, the bulk motion is the relative motion (linear and / or angular) between the scanning probe and the target tooth during this time delay. In some embodiments, this bulk motion may cause an error in the orientation and / or position of the next scanning point relative to the current scanning point. In other words, the orientation and / or position of the next scanning point may be determined not only by the beam steering module.

[0164] In some embodiments, the bulk motion information includes orientation, position, linear or angular velocity, linear or angular acceleration, etc. It can be obtained by a motion tracking device (e.g., an inertial measurement unit (IMU), an accelerometer, a gyroscope, a magnetometer, an electromagnetic navigation system, an optical navigation system, etc.). In some embodiments, the true position and orientation are calculated by obtaining motion information from the tracking device and combining the bulk motion with the motion caused by the beam steering module 501 (which can be calibrated and determined) to correct this error. In some embodiments, the goal of correcting motion artifacts is to correct the bulk motion that occurs in the volume scan data having a scan data coordinate system. In some embodiments, the bulk motion information of the motion tracking device exists within its own device coordinate system. In some embodiments, calibration can be performed during manufacturing or on-site service to obtain the transformation matrix from the device coordinate system to the scan data coordinate system.

[0165] In some embodiments, such correction can be performed on each A-scan, B-scan, or even C-scan, depending on the acceptable error specifications. In some embodiments, if the expected bulk motion is fast enough such that the motion artifacts in adjacent A-scans become unacceptable, motion correction is required for each A-scan. In some embodiments, the scanner system uses a dual galvanometer mirror beam steering system. In some embodiments, the motion tracking system obtains the position change ΔP and the orientation change Δθ of a B-scan. ΔP and Δθ are vectors. Assume that a B-scan consists of n A-scans. The bulk motion correction for each A-scan can be expressed as follows:

[0166] A-scan #1 = ΔP / n; Δθ / n

[0167] A-scan #2 = 2*ΔP / n; 2*Δθ / n

[0168] A-scan #3 = 3*ΔP / n; 3*Δθ / n…

[0169] In some embodiments, the expected overall motion is slow such that motion artifacts within a B-scan are acceptable, and assuming that the motion of all A-scans within the same B-scan is the same, motion correction can be performed for each B-scan. In some embodiments, the motion correction module 605 can analyze the motion information from the motion tracking device and determine whether there is significant motion. In some embodiments, if the motion is admissible, e.g., the linear motion error of all data points is less than 100 μm, preferably less than 20 μm, then the module can choose not to perform motion correction.

[0170] In some embodiments, the system can incorporate a video camera that can be used for motion tracking. Various video motion tracking techniques can be applied, e.g., block matching algorithms, contour tracking algorithms, cross-correlation, optical flow, feature-based methods, etc.

[0171] In some embodiments, the OCT volume dataset can be used for motion tracking by matching the consecutive surface topographies of the object being scanned. Various detection algorithms can be used to extract the surface topography. Assuming partial overlap of the topographies, various surface matching algorithms can be used for motion tracking, such as iterative closest point, coherent point drift, normal distribution transform, and their variants, etc. In some embodiments, such surface matching algorithms can be used for subsets of each volume.

[0172] The refractive index of enamel has been estimated to be close to 1.6, or similar to glass. When a light beam (including a laser beam) enters the enamel region, some of the light rays can be refracted in different directions rather than remaining parallel. They may also become shorter due to the higher refractive index. In some embodiments, the OCT sensor will receive the returned laser beam and reconstruct the resulting scan image assuming the beam remains straight. In some embodiments, to correct for this, we first use computer vision algorithms to detect the enamel surface. In some embodiments, these algorithms are detailed in the registration module 607. In some embodiments, for each A-line, the portion below the enamel surface is rescaled to remain accurate with respect to the light ray shortening effect. In some embodiments, the plane normal at each point on the enamel surface is calculated, and the local slope is calculated using the surrounding points. In some embodiments, the 3D Snell's law is used to calculate the direction of the light rays entering from the enamel surface. In some embodiments, the image segment below the enamel is rotated in the appropriate direction of ray tracing as described. Figure 7 Exemplary reconstruction results of an original OCT scan according to some embodiments are depicted. The left side is the original scan, while the right side is the reconstructed scan.

[0173] Figure 8An exemplary embodiment of the stitching and fusion module 607 is depicted. In some embodiments, the topography of the tissue surface (e.g., enamel edge) that the sampling beam first impinges on can be used for volume stitching because, before impinging, light travels in air and is not subject to image distortion caused by the imaging subject. In some embodiments, the enamel edge 801 can be detected by finding the position with the strongest backscattering or the maximum backscattering gradient, or by using one of various convolutional neural network methods for tissue boundary classification, etc.

[0174] Assuming that the topographies of two adjacent scan volumes partially overlap, various algorithms can be used to determine the overlapping region. For example, the topography can be represented by a 3D point cloud ( Figure 8 802 in), a triangular mesh, or any other 3D rendering technique. Examples of point cloud registration 803 techniques can include iterative closest point, coherent point drift, and normal distribution transform. The optimal criterion (referred to as the objective function) for matching the two topographies within the overlapping region can be the mean squared error, the mean absolute error, etc. An initial match of the two point clouds can be proposed as the starting point of the algorithm. In some embodiments, the initial match is facilitated by estimating the movement of the scanning device using a motion tracking device. In some embodiments, optical flow or other keypoint-based methods are used to estimate the movement as the initial match.

[0175] The variables used for optimization can be the linear translation and rotation of the coordinates in the two volumes, or even non-rigid coordinate transformations to compensate for hand movement artifacts. Exhaustive search, gradient descent, stochastic gradient descent, etc. can be used for optimization. In some embodiments, constraints are added to the variables to avoid overfitting. An example is that the maximum number of linear translations or rotations occurring between two consecutive scan volumes can be restricted. Other constraints can include the number of iterations of optimization, the regularization term introduced by the objective function, the stopping threshold of the objective function, etc.

[0176] In some embodiments, the matching is completed after determining the transformation matrix 804 ( Figure 8 ) of the two coordinate systems of the topography. In some embodiments, the stitching algorithm is implemented such that adjacent volumes are stitched in real time. In some embodiments, such a stitching problem can be solved partially or wholly by machine learning methods (including deep learning methods). For example, for deformable registration, deep learning or other computer vision methods (such as optical flow) can be used to find the local deformation field of the image, and non-machine learning methods can also be used. Deformable registration may be useful after the initial stitching is proposed to address small differences in the scans caused by noise or movement and to align the features inside the tooth structure.

[0177] In some embodiments, a transformation matrix 805 is applied to the 3D data. In some embodiments, refinement 806 is completed, such as removing outliers and applying deformable registration under constraints. In some embodiments, the refractive index is corrected 807 before combining two scan volumes together. Figure 9B The result of such combination is shown.

[0178] Figure 10 Depicts stitching additional volume data to an existing stitched volume dataset according to some embodiments. In some embodiments, compared with Figure 8 the algorithm is modified in terms of the volume data used for point cloud registration, which are the point cloud P3 to be stitched and the volume combined with the previously stitched volume.

[0179] In some embodiments, the OCT system can be configured in other architectures, for example, time-domain OCT, spectral-domain OCT, full-field OCT, line-field OCT, etc. Figure 11 Depicts an intraoral scanner with a full-field OCT configuration according to some embodiments.

[0180] In some embodiments, the beam from the light source 31 is collimated before entering the beam splitter 32, and the light beam is split into the sample arm 33 and the reference arm 34 at this beam splitter. In some embodiments, in order to have higher acquisition efficiency, the beam splitter 32 can be a polarization beam splitter, which incorporates two quarter-wave plates 35 and 36 in the two arms. In some embodiments, the mirror 37 in the sample arm reflects the detection beam 38 towards the subject 39 of interest.

[0181] In some embodiments, the mirror 41 in the reference arm is mounted on a motorized linear translation stage. In some embodiments, by changing the position of the translation stage along the beam propagation direction, the imaging plane 40 in the sample arm can be changed and thus optical sectioning can be achieved. In some embodiments, by stepping the mirror 41 of the reference arm through a predetermined range, volume imaging of the subject can be achieved. Another mirror 42 or multiple mirrors can be used to bend the reference beam to reduce the spatial extension of the open-air configuration. In some embodiments, the beams reflected from the two arms are recombined at the beam splitter 32 and interfere. In some embodiments, the 2D camera 43 is used to convert the optical signal into an electrical signal. The post-processing of this spatially coherent OCT signal can be similar to that of temporally coherent OCT.

[0182] In some embodiments, one of the advantages of spatially coherent OCT is that optical sectioning is achieved by snapshot, thus minimizing in-frame motion artifacts, while temporally coherent OCT may have more significant motion artifacts in the frontal plane. The size of the open-air configuration of spatially coherent OCT may be larger and the signal-to-noise ratio may be lower.

[0183] Unless otherwise defined, all terms, symbols, and other technical and scientific terms used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which the claimed subject matter pertains. In some instances, for clarity and / or ease of reference, terms with commonly understood meanings are defined herein, and the inclusion of these definitions herein should not be construed as indicating a substantial difference from the meaning commonly understood in the art.

[0184] Throughout this application, various embodiments may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as a strict limitation on the scope of the disclosure. Accordingly, the description of a range should be considered to specifically disclose all possible sub-ranges as well as the individual numerical values within that range. For example, a description of a range from 1 to 6 should be considered to specifically disclose sub-ranges such as 1 to 3, 1 to 4, 1 to 5, 2 to 4, 2 to 6, 3 to 6, etc., as well as the individual numerical values within that range, e.g., 1, 2, 3, 4, 5, and 6. This rule applies regardless of the breadth of the range.

[0185] As used in this specification and the claims, the singular forms "a," "an," and "the" include plural forms unless the context clearly dictates otherwise. For example, the term "sample" includes multiple samples, including mixtures thereof.

[0186] The terms "subject," "individual," or "patient" are generally used interchangeably herein. A "subject" can be an organism that contains expressed genetic material. The organism can be a plant, animal, or microorganism, including, for example, bacteria, viruses, fungi, and protozoa. A subject can be a tissue, cell, and progeny thereof of an organism obtained in vivo or cultured in vitro. A subject can be a mammal. The mammal can be a human. A subject may be diagnosed or suspected of being at high risk of having a disease. In some cases, a subject may not be diagnosed or suspected of being at high risk of having a disease.

[0187] As used herein, the term "about" before a number means that number plus or minus 10% of its value. The term "about" before a range means the range minus 10% of its minimum value and plus 10% of its maximum value.

[0188] As used herein, the terms "treatment" or "therapy" are used to refer to a drug or other intervention for obtaining a beneficial or desired result for a recipient. Beneficial or desired results include, but are not limited to, therapeutic and / or prophylactic benefits. A therapeutic benefit may refer to eradicating or alleviating a treated symptom or an underlying symptom. Additionally, a therapeutic benefit may be achieved by eradicating or alleviating one or more physical symptoms associated with a underlying condition, such that an improvement is observed in a subject, even though the subject may still be afflicted with the underlying condition. Prophylactic effects include delaying, preventing, or eliminating the onset of a disease or disorder, delaying or eliminating the occurrence of symptoms of a disease or disorder, slowing, halting, or reversing the progression of a disease or disorder, or any combination of these effects. For a prophylactic benefit, a subject at risk of developing a particular disease or reporting one or more physical symptoms of a disease may be treated, even though the disease may not have been diagnosed.

[0189] The section headings used herein are for organizational purposes only and should not be construed as limiting the subject matter described.

[0190] I. Exemplary Embodiments

[0191] Figures 11 to 18Illustrates an exemplary embodiment of the scanning probe 1600. In some embodiments, the scanning probe 1600 includes a galvanometer optical scanner mount (galvanometer mount) 1611 located at the proximal end of the scanning probe. In some embodiments, the galvanometer mount receives a first galvanometer scanner 1615 and a second galvanometer scanner 1617. In some embodiments, the galvanometer mount 1611 further includes a movable mount 1621. In some embodiments, the movable mount 1611 allows the scanning probe to be mounted to a robotic arm. In some embodiments, the movable mount 1611 allows the scanning probe to be mounted to a handle to provide a hand-held OCT scanning system as described herein. In some embodiments, the galvanometer mount 1611 includes at least one galvanometer. In some embodiments, the galvanometer mount 1611, the first galvanometer scanner 1615, the second galvanometer scanner 1617, and at least one galvanometer mirror are referred to herein as a beam steering module. In some embodiments, the sample arm light is collimated and directed into the beam steering module. The beam steering module may have two mirrors as beam reflecting elements. In some embodiments, the mirrors are mounted on oscillating components (such as two galvanometer motors), which form a galvanometer mirror subsystem. The galvanometer mirror subsystem can be used for its determined motion feedback and control as well as scanning linearity, which may be required for simplified data reconstruction in subsequent processing stages. The oscillation axes of the two mirrors can be perpendicular to facilitate position control, such that one mirror controls the frequency sweep of the light beam in one direction, while the other mirror controls the frequency sweep in a direction orthogonal thereto. In some embodiments, the galvanometer mirror subsystem is configured to scan not only as fast as possible, but also with minimal scanning field distortion and high reproducibility. Thus, the subsystem can be configured for one-way scanning and only use the linear scanning portion.

[0192] In some embodiments, the beam steered by the galvanometer mirror subsystem is then refocused by a lens 1626, reflected by a mirror 1637, and exits through an imaging window 1607. In some embodiments, a transparent window is mounted to prevent contamination of the lens system.

[0193] In some embodiments, the galvanometer mount 1611 is coupled to the aligner 1609 via an adapter 1613. In some embodiments, the aligner houses a dichroic mirror 1639. In some embodiments, the video camera 1625 is coupled to the aligner 1609. In some embodiments, the dichroic mirror 504 is configured to transmit OCT light and reflect visible light. In some embodiments, the visible light is reflected or scattered by the tooth and redirected to the camera 1625. In some embodiments, the camera lens system 1627 is provided for the camera 1625 and attached to the aligner 1609. In some embodiments, the camera lens system 1627 is configured to match the focal plane and the depth of focus of the OCT beam profile. In some embodiments, the aligner 1609 further includes a lid 1619. The lid 1699 may facilitate access to the dichroic mirror 1639 housed within the aligner 1609. In some embodiments, a lens 1626 is coupled to the aligner. In some embodiments, the OCT lens system includes the lens 1626, a first angular spacer 1629, a window 1631, a second angular spacer 1632, and a lens tube 1633. In some embodiments, the lens tube 1633 houses components of the OCT lens system.

[0194] In some embodiments, a cannula 1601 is provided at the distal end of the scanning probe 1600. In some embodiments, the cannula 1601 is coupled to the aligner and houses components of the OCT lens system. In some embodiments, the cannula 1601 includes a mirror 1637 to direct the sample arm through the imaging window 1607. In some embodiments, the imaging window includes a transparent cover to prevent contamination of the lens system.

[0195] In one aspect, the present disclosure provides an optical coherence tomography system that performs fast, sparse scans and stitches together multiple such scans to generate a dense image. As described above, traditional OCT scanning techniques may have several limitations that limit their use in dentistry, such as limited penetration depth, a small field of view (FOV) that prevents full-arch imaging, a long acquisition time that may result in motion artifacts within a single volume, and the need for complex registration to achieve the surface fidelity required for an intraoral scanner (IOS) or to guide automated tooth preparation procedures. The present disclosure can address the above limitations by providing improved OCT scanning systems and methods. For example, an OCT scanner according to the present invention can advantageously avoid an unacceptable amount of motion blur by quickly traversing its corresponding scan pattern, typically completing an entire two-dimensional frame faster than it takes a traditional raster scanner to complete one raster line segment. To quickly traverse its scan pattern, the systems and methods herein can provide a sparse sampling method by reducing the number of A-scans per unit length of the scan pattern. For example, fewer A-scans can be made per unit length of the scan pattern compared to a traditional OCT scanner as described above. To compensate for the sparsity of the sampling points along the corresponding scan segments and for the gaps between the corresponding segments of the trajectory, these embodiments acquire and combine several partially overlapping frames for each study.

[0196] Optical coherence tomography (OCT) is an imaging technique that uses low-coherence light (typically near-infrared light) to capture two-dimensional and three-dimensional images with micron-scale resolution from within an optically scattering medium such as biological tissue. OCT is based on low-coherence interferometry. In traditional long-coherence length interferometry (i.e., laser interferometry), the interference of light occurs over distances on the order of meters. However, in OCT, due to the use of a broadband light source (i.e., a light source that emits light over a broad wavelength range, such as a superluminescent light-emitting diode and a laser with ultrashort pulses (femtosecond laser)), this interference effect is reduced to micron-scale distances.

[0197] Figure 19 is a schematic block diagram of an optical coherence tomography system 100 according to the prior art. The light in the OCT system 100 is split into two arms: a sample arm 102 containing the structure 104 to be measured, and a reference arm 106 typically containing a mirror 108. The combination of the reflected light from the sample arm 102 and the reference light from the reference arm 106 produces an interference pattern, provided that the light from both arms 102 and 106 travels an equal optical distance (i.e., a distance difference less than the coherence length of the light). By scanning the mirror 108 in the reference arm 106, reflectivity profiles of the structure 104 at different depths can be obtained. The interference amount is proportional to the amount of reflected light, which enables the distinction of portions of the structure 104 to be measured having different reflectivity characteristics at different depths. Any light beyond the short coherence length range does not interfere, thus enabling OCT to detect the specific depth of the structure 104 to be measured and generate a reflectivity profile. This reflectivity profile is called an A-scan and contains information about the spatial dimensions and structure within the structure of interest. An A-scan (axial or depth scan) represents data recovered by conceptually drilling individual "holes" at different depths into the structure 104 to be measured.

[0198] Cross-sectional tomography (B-scan) can be achieved by laterally combining a series of these axial depth scans (A-scans) or by scanning a mirror 110 in the sample arm 102. For example, scanning the mirror 110 in one dimension moves the light in the sample arm 102 so as to project the light onto successive points along a line segment on the surface of the structure 104 to be measured (when looking down along the axis 112 of the sample arm 102). Figure 20 The line 200 in [reference] shows a line segment on the surface of the structure 104 to be measured. Modern OCT systems sample individual points on the line 200, thereby generating individual pixels (not shown) along the line 200.

[0199] The OCT system can combine multiple A-scans along a B-scan to generate a two-dimensional reflectivity profile, such as Figure 21 shown in 2100 in [reference]. The plane of the two-dimensional reflectivity profile 2100 represents a "slice" taken through the structure 104 to be measured. Figure 21 The top line 2102 (shown in thick line) of the two-dimensional reflectivity profile 2100 in [reference] corresponds to Figure 20 the line 200 in [reference].

[0200] The OCT system can combine multiple A / B-scans, such as Figure 22 the scans 2200, 2202, 2204, 2206, and 2208 represented in [reference], to generate a three-dimensional reflectivity profile 2210. Figure 22 The top thick line in [reference] is represented by the thick line 2212 and corresponds to Figure 21 the line 2100 in [reference] and Figure 20 the line 200 in [reference].

[0201] A conventional method of performing a combined B / C scan is to project a reference arm beam along a grating 2300 onto successive points towards a structure 104 under test, as Figure 23 shown. The grating is characterized by a plurality of parallel spaced scan line segments. Figure 23 The retrace line segments are shown in dashed lines and the scan direction is shown by arrows. However, the grating may not have retrace line segments, depending on how the light of the reference arm is redirected. For example, two rotating polygonal mirrors can generate a grating without retrace.

[0202] One complete grating scan (i.e., one traversal of the pattern) is the same as that shown in Figure 23 , which is referred to as a frame. The area covered by one traversal of the pattern is referred to as the outer boundary or bounding box of the two-dimensional scan area, as illustrated by the outer boundary 2302 of the two-dimensional scan area.

[0203] Unfortunately, conventional intraoral OCT scanning methods and devices are too slow, especially for hand-held scanning wands. A conventional single-frame grating scan will take several seconds to complete, during which time the system will produce an unacceptable amount of motion blur due to accidental movement of the operator's hand. For example, in an exemplary scan pattern 2400 as shown in Figure 24 , accidental movement of the scanning wand may result in non-uniform spacing between successive grating line segments, x-distortion and / or y-distortion of the grating pattern that distorts the outer boundary 2402 of the two-dimensional scan area in the x and / or y directions, and bending of individual grating line segments (not shown). In addition, the degree of motion blur cannot be estimated because a single grating pattern does not have sampling points where two grating lines cross. Thus, there is no common reference point for two grating line segments to detect the same point on the structure under test, and thus accidental movement cannot be detected or measured. Some prior art systems include additional cameras for detecting accidental movement, but such systems are bulky, expensive and inconvenient to use.

[0204] Conventional OCT scan frames typically cover too small an area to image a larger tooth or a larger part of the oral cavity. Thus, it may be necessary to combine multiple OCT scan frames to form a single image of the structure under test. However, it is extremely challenging or impossible for an operator to keep the scanning wand stationary for each such frame and only move the wand controllably between frames.

[0205] The systems and methods of the present disclosure can address these and other problems associated with the above-described conventional OCT scanning techniques. Specifically, the OCT systems and methods herein can advantageously avoid unacceptable amounts of motion blur by traversing their respective scan patterns in a short period of time (e.g., completing an entire two-dimensional frame much faster than the time it typically takes a conventional raster scanner to complete one raster line segment). In some cases, compared to conventional OCT scanners, the scan time can be shortened by performing fewer A-scans per unit scan pattern length. Compared to conventional scanners that typically perform "dense" scans, the systems and methods herein are capable of performing "sparse" scans without compromising imaging quality. As Figure 20 described, conventional OCT systems sample individual points on line 200, thereby generating individual pixels (not shown) along line 200. Figure 25 Shown is a hypothetical dense scan of line segment 700 by a conventional OCT system consisting of multiple individual sampling points (pixels), and a hypothetical sparse scan of line segment 702 of one embodiment of the present disclosure. Note that the sampling points on line segment 700 are positioned more densely than the sampling points on line segment 702, although Figure 25 the specific ratio of sample spacing shown is for illustrative purposes only.

[0206] Some embodiments of the present disclosure employ curved scan patterns (such as Lissajous figures) or spiral scans. Figure 26 Shown is an exemplary Lissajous figure 2600. The Lissajous figure can be the same as the Lissajous scan path described above.

[0207] As schematically shown in FIG. 9, a tomographic system 900 according to an embodiment of the present invention includes an OCT system 902 having an optical detector 903, a probe housing 904 of a scanning wand, and a movable mirror system 906 disposed within the housing. The mirror system 906 is configured to redirect (such as by reflection) a portion of the sample arm 908 of the OCT system 902. The redirected portion 910 of the sample arm 908 extends through a window 914 in the probe housing 904 into free space 912 outside the probe housing 904. A controller 916 drives a motor 918 to repeatedly change the orientation of the mirror system 906 relative to two different axes (axes illustrated by 920 and 922), thereby repeatedly scanning the surface 924 of the structure under test 926 on a structure along a trajectory 928 of a determined, smooth two-dimensional scan pattern (illustrated by the Lissajous pattern 930 shown in the inset of the drawing) using the light of the sample arm 908. If the motor 918 oscillates the mirror system 906 relative to two orthogonal axes 920 and 922 according to corresponding sine wave signals, a suitable Lissajous pattern can be achieved based on the relative frequencies and phases of the sine wave signals. Other exemplary Lissajous figures are in Figure 10is shown above. As described above, other scan patterns may be used.

[0208] Temporarily return Figure 26 , the dashed line 2602 represents the elastic boundary shape surrounding the Lissajous figure 2600. As used herein, the term outer boundary of the two-dimensional scan region refers to a single closed-loop elastic boundary shape that tightly fits to surround the entire scan pattern when viewed downward along the axis 932 of the sample arm 908 ( Figure 27 ), i.e., when projected onto the surface 924 of the structure 926, as exemplified by the dashed line 2602. When viewed from a perspective outside the window 914, due to the topography of the surface 924, the trajectory of the points on the anatomical structure 926 illuminated by the sample arm 908 during scanning may have a shape different from the scan pattern.

[0209] Although in each frame, the tomographic system 900 in FIG. 9 scans multiple portions of the structure 926 to be measured within the outer boundary of the two-dimensional scan region, each frame includes only a subset of the samples from the region outside the outer boundary of the two-dimensional scan region. Specifically, each frame includes samples acquired along a line (usually a curved and possibly self-crossing line such as the Lissajous figure 930) corresponding to the trajectory 928 of the sample arm beam. Thus, during the scanning of each frame, multiple portions of the structure 926 remain unsampled. For example, the regions 934 and 936 (see the inset in Figure 27 ) are not sampled by the Lissajous figure 930. The trajectory 928 (especially a complex trajectory such as the Lissajous figure 930) can be considered to be composed of multiple line segments that are not necessarily straight lines, as exemplified in Figure 29 . Figure 29 The Lissajous figure 930 is mainly shown in dashed lines. However, for clarity, the exemplary line segments 1100, 1102, 1104, and 1106 are shown in solid lines. These line segments can have any length. The line segments do not necessarily start and end at the intersection points with other lines or line segments of the trajectory.

[0210] Each traversal of the scan pattern defines multiple gaps between the corresponding line segments of the trajectory. As described above, these gaps (exemplified by the regions 934 and 936) are not illuminated by the light of the sample arm 908 during the traversal.

[0211] To compensate for the sparsity of the sampling points on their corresponding scan line segments as discussed in reference Figure 25 and to compensate for the gaps between the corresponding line segments of the trajectory as discussed in reference Figure 27 , embodiments of the present disclosure acquire and combine several partially overlapping frames during each imaging, as schematically shown in Figure 30 . For simplicity of illustration, Figure 26 , Figure 27 and Figure 29The Lissajous figure in [it] is used as a scanning pattern. However, other scanning patterns can be used, as discussed in more detail herein. Each such frame is acquired from a different perspective of the structure under test, as Figure 31 illustrated in [it]. These viewpoints are generated by an operator or machine translating the scanning wand 1300 along a path 1302 in space, which path is close to the structure under test 1304 on the structure, such as a tooth of a living patient. Thus, embodiments of the present invention utilize the movement of the scanning wand 1300, rather than attempting to suppress the movement as in the prior art.

[0212] Figure 30 illustrates exemplary successive traversals 1200 of a scanning pattern 930( Figure 27 ). 27), including a first traversal 1202, a second traversal 1204, a third traversal 1206, a fourth traversal 1208, and a fifth traversal 1210. For clarity of the drawings, each traversal 1202 to 1210 is shown in a different dashed line type. Although these five traversals 1202 to 1210 are shown, the imaging process can consist of any number of traversals. As described above, the corresponding scans are performed at different positions along a path 1303( Figure 31 ) of the scanning wand 1300. In the example shown in Figure 30 , the x coordinate and y coordinate of the upper left corner of each successive scan gradually increase compared to the previous scan. The outer boundary of the corresponding scan area of each scan ( Figure 30 not shown in [it]) partially overlaps the outer boundary of the scan area of at least one other such scan. Successive traversals of the scanning pattern illuminate corresponding portions of at least some of the gaps defined by at least one other such traversal of the scanning pattern. For example, the second traversal 1204 (such as the portion of the second traversal denoted by reference numeral 1212) illuminates a portion of the gap (denoted by reference numeral 1214) defined by the first traversal 1202.

[0213] For each traversal, the controller 916( Figure 27 ) receives pixel image data regarding a corresponding portion of the surface 924 of the anatomical structure 926 from the optical detector 903. The pixel image data for each traversal 1202 to 1210( Figure 30 ) contains a first number of pixels. However, the successive traversals 1202 to 1210 "fill" portions of some of the gaps of the other traversals among the traversals 1202 to 1210, that is, are interwoven with some of the gaps of the other traversals. Thus, for successive traversals, the controller 916 stitches together the pixel image data of the successive traversals 1202 to 1210, thereby generating a stitched surface image with a number of pixels greater than the first number of pixels. Any conventional two-dimensional or three-dimensional point cloud registration and stitching method can be used.

[0214] Although the number of pixels generated by controller 916 is greater than the mosaicked surface images of each individual traversal 1202-1210, the controller does not rely on interpolation methods to achieve this increase in pixel density. Instead, a compressive sensing scheme such as that described in U.S. Patent No. 11,497,402 uses random or pseudo-random samples to reconstruct the surface. When using compressive sensing type data acquisition, a large class of sparse signals can be reconstructed by employing a sub-Nyquist sampling rate, which contain a small number of dominant components in certain domains. Different from the uniformly spaced signal measurements based on Nyquist sampling, compressive sensing theory proves that several types of uniformly random sampling protocols can achieve successful reconstruction with high probability.

[0215] U.S. Patent No. 11,497,402 discloses random sampling OCT in two spatial dimensions x and y. The x scan positions are generated by a first pseudo-random sequence, and the y scan positions are generated by a second pseudo-random sequence. The two-dimensional sampling grid is determined by interleaving the x and y sequences. The sampling tuple (x i , y i ) is created from the x w components of the x random sampling sequence x = {x1, x2,..., x i} and the y D components of the y random sampling sequence y = {y1, y2,..., y i}. This forms a random or pseudo-random spacing arrangement, which helps to reduce the number of samples required to generate an OCT reconstruction.

[0216] In another aspect, embodiments of the present invention do not rely on random or pseudo-random spacing arrangements. As used herein, the term "determined" refers to non-random and non-pseudo-random. As described above, controller 916 drives motor 918 to repeatedly change the orientation of mirror system 906 relative to two different axes, so as to repeatedly scan the surface 924 of the structure 926 to be measured on the structure along the trajectory 928 according to a determined scan pattern by using the light of sample arm 908.

[0217] Some embodiments use a stitched surface image as a map for stitching together voxel subsurface data from the optical detector 903. As used herein, a pixel represents an image element of a one-dimensional or two-dimensional image or a voxel (volume element of a three-dimensional image). For each pass, the controller 916 receives voxel subsurface data from the optical detector 903 regarding a corresponding subsurface portion of the anatomical structure 926. The voxel subsurface data for each pass includes a second number of voxels. As in the case of surface stitching, successive passes 1202 to 1210 "fill" portions of some of the gaps of other passes among passes 1202 to 1210 (interleave). Thus, for multiple successive passes, the controller 916 stitches together the voxel subsurface data of the multiple successive passes, thereby generating a stitched subsurface three-dimensional volume image with a number of voxels greater than the second number of voxels.

[0218] In some embodiments, the wavelength of the sample arm of the OCT 902 is approximately 1310 nm, the A-scan rate is approximately 200 kHz, the OCT transverse beam spot size is approximately 35 μm, and the imaging range is approximately 16 mm. The field of view is approximately 8 mm × 8 mm.

[0219] Applicability of raster scanning

[0220] Although a smooth scan pattern is described herein, some embodiments employ raster scanning and partially overlap (interleave) with the outer boundary of the two-dimensional scan region, as referenced Figure 30 described. The raster lines of successive frames are interleaved to illuminate at least part of the gaps defined by the previous raster frame.

[0221] Slight discontinuities in the scan pattern

[0222] Although a smooth scan pattern is described herein, in some embodiments, the controller is configured to drive the motor to repeatedly change the orientation of the mirror system relative to two different axes, thereby repeatedly scanning the surface of the anatomical structure with the light of the sample arm along a trajectory according to the scan pattern, wherein at least 80% of the scan pattern along the trajectory is smooth. For example, the scan pattern can be a spiral, scanning back from the center to the outer edge / from the outer edge back to the center.

[0223] Such a system can be implemented using two galvanometric mirrors or a single mirror (such as a MEMS mirror) that can be redirected in two dimensions.

[0224] Motion detection using cross traversals of line segments

[0225] The position where one line segment of the scan pattern crosses (intersects) another line segment of the same pass of the scan pattern can provide information to quantify the housing 904 ( Figure 27)Motion at two time points. For example, the intersection point 1216 formed by two line segments 1218 and 1220 traversing 1204 ( Figure 30 ) can be used to quantify the motion of the housing 904 between (a) the time when the beam of the sample arm 908 passes through the line segment 1218 near the intersection point 1216 and (b) the time when the beam of the sample arm 908 passes through another line segment 1220 near the intersection point 1216.

[0226] Ideally, if the housing 904 does not move between times (a) and (b), the beam of the sample arm 908 should detect the same or very similar regions of the structure 926 under test near the intersection point 1216. However, if the controller 916 detects a significant difference (e.g., greater than a predetermined amount) between the portions of the structure 926 under test detected by the beam at times (a) and (b), the controller 916 may conclude that the viewing point of the housing 904 has changed significantly between times (a) and (b). Optionally, the controller 916 may discard the current frame due to excessive motion blur.

[0227] Alternatively, the controller 916 can estimate the amount of change in the field of view by analyzing the differences between the portions of the structure 926 under test detected by the beam at times (a) and (b). For example, based on information about the characteristics (such as reflectivity, density, or color) of the portions of the structure 926 under test at times (a) and (b) and the expected spatial gradient of the characteristics of the structure 926 under test, the controller 916 can estimate the spatial distance between the detected positions of the two samples and thus estimate the amount or rate of translation of the housing 904. The expected spatial gradient of the characteristics can be a pre-programmed assumption, or it can be a user input value, or the controller 916 can automatically estimate the gradient based on other samples of the structure 926 under test.

[0228] Definitions

[0229] As used herein, unless the context otherwise indicates, the following terms shall have the following meanings.

[0230] "Continuously" means continuously or repeatedly, although not necessarily permanently. The term "continuously" encompasses periodically and occasionally. Continuously generating a signal means generating a signal that varies continuously over time, or means generating a series (more than one) of discrete signals over time. Continuously generating a value (such as an error value) means generating a value that varies continuously (such as an analog value represented by a continuously varying voltage), or means generating a series (more than one) of discrete values over time, such as a series of numerical or analog values.

[0231] Although the present invention has been described by way of the above-described exemplary embodiments, modifications and variations can be made to the illustrated embodiments without departing from the inventive concept disclosed herein. For example, although specific parameter values (such as materials and dimensions) may be described with respect to the disclosed embodiments, within the scope of the present invention, all parameter values can vary within a wide range to suit different applications. Unless the context otherwise indicates or will be understood by those skilled in the art, a term such as "about" means within ±20%.

[0232] As used herein (including in the claims), the term "and / or" used in conjunction with a list of items means one or more of the items in the list, i.e., at least one item in the list, but not necessarily all items in the list. As used herein (including in the claims), the term "or" used in conjunction with a list of items means one or more of the items in the list, i.e., at least one item in the list, but not necessarily all items in the list. "Or" does not mean "exclusive or".

[0233] As used herein (including in the claims), an element described as configured to perform one operation "or" another operation satisfies an element configured to perform only one of the two operations. That is, the element is not necessarily configured to perform one operation in one mode in which the element performs one operation and perform another operation in another mode in which the element performs the other operation. However, the element can but does not have to be configured to perform more than one operation.

[0234] Although various aspects of the embodiments may be described with reference to flowcharts and / or block diagrams, the functions, operations, decisions, etc. of all or part of each block diagram or a combination of block diagrams can be combined, separated into individual operations, or performed in other orders. References to "modules", "operations", "steps" and similar terms are for convenience and are not intended to limit their embodiments. All or part of each block, module, operation, step or combination thereof can be implemented as computer program instructions (such as software), hardware (such as combinational logic, application specific integrated circuit (ASIC), field programmable gate array (FPGA), processor or other hardware), firmware or a combination thereof.

[0235] The controller 916 etc. or a part thereof can be implemented by one or more suitable processors that execute instructions stored in a memory or are controlled by instructions stored in a memory. Each processor can be a general-purpose processor, such as a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a dedicated processor, etc., as the case may be, or a combination thereof.

[0236] The memory can be a random access memory (RAM), a read-only memory (ROM), a non-volatile memory (NVM), a non-volatile random access memory (NVRAM), a flash memory, or any other memory, or a combination thereof, suitable for storing control software or other instructions and data. The instructions defining the functions of the present invention can be transmitted to the processor in various forms, including but not limited to, information permanently stored on a tangible non-transitory non-writable storage medium (e.g., a read-only memory device inside a computer such as a ROM, or a device readable by a computer I / O accessory such as a CD-ROM or a DVD), information variably stored on a tangible non-transitory writable storage medium (e.g., a floppy disk, a removable flash memory, and a hard disk), or information transmitted to a computer through a communication medium (including a wired or wireless computer network). Additionally, although embodiments may be described in conjunction with various illustrative data structures, database schemas, etc., the present system can be implemented using different data structures, schemas, etc.

[0237] The disclosed aspects or parts thereof can be combined in ways not listed and / or not explicitly claimed herein. Additionally, the embodiments disclosed herein can be suitably implemented without any element not specifically disclosed herein. Thus, the present invention should not be regarded as limited to the disclosed embodiments.

[0238] As used herein, label terms such as "first", "second", and "third" are used to distinguish corresponding elements from each other, such as mirrors or traversals, and are not intended to represent any particular order or total number of mirrors or traversals in any particular embodiment. Thus, for example, a given embodiment may include only a second mirror and a third traversal.

[0239] Although the preferred embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that these embodiments are provided by way of example only. Many variations, changes, and substitutions will occur to those skilled in the art without departing from the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein can be used to practice the present invention. The appended claims are intended to define the scope of the present invention, and the methods and structures within the scope of these claims and their equivalents are thereby covered.

Claims

1. A tomographic scanning system, comprising: A probe housing that defines a window and is configured to translate along a path proximate to an anatomical structure within a living patient, the anatomical structure having a surface; An optical coherence tomography system including a sample arm and an optical detector, wherein a portion of the sample arm extends through the window into free space outside the probe housing; A movable mirror system disposed within the probe housing and configured to redirect the sample arm; A motor disposed within the probe housing and coupled to the mirror system; and A controller configured to automatically: Drive the motor to repeatedly change the orientation of the mirror system relative to two different axes so as to repeatedly scan the surface of the anatomical structure with light from the sample arm along a trajectory according to a determined, smooth two-dimensional scan pattern such that: On a respective portion of the surface of the anatomical structure, each traversal of the scan pattern defines an outer boundary of a respective two-dimensional scan region; Each traversal of the scan pattern defines a plurality of gaps between respective line segments of the trajectory, wherein the gaps are not illuminated by the light during that traversal; and In successive traversals of the scan pattern, respective scans are performed from different positions along the path, and the outer boundary of the respective scan region of each scan in the respective scans partially overlaps the outer boundary of the scan region of at least one other scan in the respective scans to illuminate a respective portion of at least some of the gaps defined by at least one other such traversal of the scan pattern; For each traversal, receive pixel image data regarding the respective portion of the surface of the anatomical structure from the optical detector, wherein the pixel image data for each traversal includes a first number of pixels; and For a plurality of successive traversals, wherein respective scans are performed from different positions along the path, and the outer boundary of the respective scan region of each scan in the respective scans partially overlaps the outer boundary of the scan region of at least one other scan in the respective scans, stitch together the pixel image data of the plurality of successive traversals to generate a stitched surface image having a number of pixels greater than the first number of pixels.

2. The computed tomography system according to claim 1, wherein, The controller is configured to analyze image features of the pixel image data of the plurality of successive traversals to estimate a respective displacement of one of the respective scan region outer boundaries from another of the respective scan region outer boundaries.

3. The tomographic scanning system according to claim 1, wherein, The scan pattern includes a Lissajous figure.

4. The tomographic scanning system according to claim 1, wherein, The scan pattern is anisotropic.

5. The tomographic scanning system according to claim 1, wherein, The controller is configured to automatically: For each traversal, receive voxel subsurface data regarding the respective subsurface portion of the anatomical structure from the optical detector, wherein the voxel subsurface data for each traversal includes a second number of voxels; and For the plurality of successive traversals, stitch together the voxel subsurface data of the plurality of successive traversals to generate a stitched subsurface three-dimensional volume image having a number of voxels greater than the second number of voxels.

6. The tomographic scanning system according to claim 5, wherein, The controller is configured to automatically: Use the pixel image data to detect the enamel / air boundary of the anatomical structure; Estimate the amount of refraction of the light at the enamel / air boundary based on the difference between the refractive index of enamel and the refractive index of air; And Change the coordinates of the voxel subsurface data to at least partially compensate for the refraction of the light at the enamel / air boundary.

7. The tomographic scanning system according to claim 5, wherein, The controller is configured to automatically: Use the voxel subsurface data to detect the enamel surface of the anatomical structure; Use the voxel subsurface data to detect the enamel / dentin boundary within the anatomical structure; Estimate the thickness of the enamel between the enamel surface and the enamel / dentin boundary; Estimate the amount of refraction of the light within the enamel based on (a) the thickness of the enamel and (b) a predetermined refractive index of the enamel; And Change the coordinates of the voxel subsurface data to at least partially compensate for the refraction of the light within the enamel.

8. The tomographic system according to claim 1, wherein, The mirror system has a resonance frequency; and The controller is configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 50% of the resonance frequency.

9. The tomographic scanning system according to claim 8, wherein, The controller is configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 30% of the resonance frequency.

10. The tomographic scanning system according to claim 8, wherein, The controller is configured to drive the motor to repeatedly change the orientation of the mirror system at a frequency within 20% of the resonance frequency.

11. A tomographic system, comprising: A probe housing that defines a window and is configured to translate along a path adjacent to an anatomical structure within a living patient, the anatomical structure having a surface; An optical coherence tomography system including a sample arm and an optical detector, wherein a portion of the sample arm extends through the window into free space outside the probe housing; A movable mirror system disposed within the probe housing and configured to redirect the sample arm; A motor disposed within the probe housing and coupled to the mirror system; and A controller configured to automatically: Drive the motor to repeatedly change the orientation of the mirror system relative to two different axes, thereby repeatedly scanning the surface of the anatomical structure with the light of the sample arm along a trajectory according to a determined two-dimensional scan pattern such that: On a corresponding portion of the surface of the anatomical structure, each traversal of the scan pattern defines an outer boundary of a corresponding two-dimensional scan region; Each traversal of the scan pattern defines a plurality of gaps between corresponding line segments of the trajectory, wherein the gaps are not illuminated by the light during that traversal; and In consecutive traversals of the scan pattern, corresponding scans are performed from different positions along the path, and the outer boundary of the corresponding scan region of each scan of the corresponding scans partially overlaps the outer boundary of the scan region of at least one other scan of the corresponding scans to illuminate at least some corresponding portions of the gaps defined by at least one other such traversal of the scan pattern; For each pass, receive pixel image data from the optical detector regarding a corresponding portion of the surface of the anatomical structure, where the pixel image data for each pass includes a first number of pixels; and For a plurality of consecutive passes, where corresponding scans are performed from different positions along the path and an outer boundary of a corresponding scan area of each scan in the corresponding scans partially overlaps an outer boundary of a scan area of at least one other scan in the corresponding scans, stitch together the pixel image data of the plurality of consecutive passes to generate a stitched surface image having a number of pixels greater than the first number of pixels.

12. The tomographic scanning system according to claim 11, wherein, The scan pattern includes a grating.

13. A tomography system, comprising: A probe housing that defines a window and is configured to translate along a path proximate to a surface of an anatomical structure within a living patient; An optical coherence tomography system including a sample arm and an optical detector, where a portion of the sample arm extends through the window into free space outside the probe housing; A movable mirror system disposed within the probe housing and configured to redirect the sample arm; A motor disposed within the probe housing and coupled to the mirror system; and A controller configured to automatically: Drive the motor to repeatedly change the orientation of the mirror system relative to two different axes so as to repeatedly scan the surface of the anatomical structure with light from the sample arm along a trajectory according to a scan pattern, where at least 80% of the scan pattern along the trajectory is smooth, and repeatedly scanning the surface of the anatomical structure such that: On a corresponding portion of the surface of the anatomical structure, each pass of the scan pattern defines a corresponding outer boundary of a two-dimensional scan area; Each pass of the scan pattern defines a plurality of gaps between corresponding line segments of the trajectory, where the gaps are not illuminated by the light during that pass; and In consecutive passes of the scan pattern, corresponding scans are performed from different positions along the path and an outer boundary of a corresponding scan area of each scan in the corresponding scans partially overlaps an outer boundary of a scan area of at least one other scan in the corresponding scans to illuminate a corresponding portion of at least some of the gaps defined by at least one other such pass of the scan pattern; For each pass, receive pixel image data from the optical detector regarding a corresponding portion of the surface of the anatomical structure, where the pixel image data for each pass includes a first number of pixels; and For a plurality of consecutive passes, where corresponding scans are performed from different positions along the path and an outer boundary of a corresponding scan area of each scan in the corresponding scans partially overlaps an outer boundary of a scan area of at least one other scan in the corresponding scans, stitch together the pixel image data of the plurality of consecutive passes to generate a stitched surface image having a number of pixels greater than the first number of pixels.

14. The tomographic scanning system according to claim 13, wherein, The scan pattern includes a helix.

15. The tomography system according to claim 1, wherein The mirror includes a first mirror and a second mirror, and the motor is configured to continuously change the orientation of the first mirror along a first axis and continuously change the orientation of the second mirror along a second axis different from the first axis.

16. The tomographic scanning system according to claim 1, wherein, The mirror and the motor together include a biaxial microelectromechanical system.

17. The tomographic system according to claim 1, further comprising: a memory storing calibration data characterizing the optical non-ideality of each tomographic system; and wherein: the controller is configured to modify the data received from the optical detector to at least partially compensate for the optical non-ideality.

18. The tomographic scanning system according to claim 17, wherein, The optical non-ideality includes at least one of the following: lens aberration, deformation of the mirror caused by driving of the motor, and optical misalignment.

19. The tomographic scanning system according to claim 1, wherein, The controller is configured to drive the motor to change the orientation of the mirror system along the two axes, so as to repeatedly scan the structure along a first closed-loop two-dimensional scan trajectory and a second closed-loop two-dimensional scan trajectory, wherein the first closed-loop two-dimensional scan trajectory provides more sampling points than the second closed-loop two-dimensional scan trajectory.

20. The tomographic system according to claim 19, further comprising: a motion detector mechanically coupled to the probe housing and configured to detect the motion of the probe housing; and the controller is configured to automatically: detect whether the motion of the probe housing is less than a predetermined value; control the motor such that: when the motion of the probe housing is less than the predetermined value, the motor changes the orientation of the mirror to scan the teeth along the first closed-loop two-dimensional scan trajectory; and when the motion of the probe housing is not less than the predetermined value, the motor changes the orientation of the mirror to scan the teeth along the second closed-loop two-dimensional scan trajectory.

21. The tomographic system according to claim 1, further comprising a pulse-to-continuous-wave optical buffer, the pulse-to-continuous-wave optical buffer including: a laser configured to output a series of pulses, each pulse having a pulse width; an N-way optical beam splitter coupled to the output of the laser, wherein N>1; at least N-1 delay lines, with the respective inputs of each of the at least N-1 delay lines coupled to the respective outputs of the N-way optical beam splitter; and an N-way optical combiner, wherein: the respective outputs of each delay line are coupled to the respective inputs of the N-way optical combiner; and each delay line is configured to impart a delay equal to a different integer multiple of the pulse width plus a constant k (k≥0).

22. The tomographic system according to claim 21, wherein: the laser is configured to output light according to a duty cycle (D); and N=(1 / D)-1.

23. The tomographic system according to claim 21, wherein: the laser is configured to output light according to a duty cycle (D); and N=1 / D.

24. The tomographic system according to claim 21, further comprising: a polarization detector optically coupled to the output of the N-way optical combiner; and a polarization controller, the polarization controller: Optical coupling is between an input of the N-way optical combiner and the laser, communicatively coupled to the polarization detector, and the polarization controller is configured to adjust the polarization of the light passing through the polarization controller to match the polarization of the light delivered to another input of the N-way optical combiner.

Images