WIDEFIELD SWEPT SOURCE OCT AND METHODS FOR MOVING OBJECTS

Description

[0001] The invention relates to a wide-field swept-source OCT and method for moving objects, in particular the anterior chamber of the human eye.

[0002] Ophthalmic OCT systems have been used for many years to measure the human retina and also structures in the anterior chamber of the eye. A well-known anterior chamber OCT system was developed by Carl Zeiss in the form of a confocal scanner.

[0003] For anterior chamber measurements, OCT systems developed for retinal imaging are often adapted with an add-on lens. Because the imaging properties for anterior chamber measurements differ from those for retinal imaging, and OCT systems are optimized for the retina, imaging performance in the anterior chamber is not optimal. The key differences in imaging these two tissues are as follows: 1. All beams reaching the retina are clipped by the iris, so that OCT systems designed for the retina all operate with a similar pupil size of 1.0 to 1.3 mm and resulting numerical apertures of 0.04 at their maximum. This limits the lateral resolution, but the depth of field, which determines the accessible depth range during measurement, remains sufficient for imaging the retina down to the measurement depth of just under one millimeter, which is limited by scattering. 2. The tissues in the retina, especially the retinal pigment epithelium and the choroid behind it, are highly scattering. To minimize the resulting multiple scattering, most OCT systems are designed as fully confocal systems.Since the OCT illumination and detection waves are imaged by the human eye lens, which exhibits certain aberrations, only a slightly better lateral resolution can be achieved even in very small image fields and depth ranges. 3. The tissues in the anterior chamber (cornea, aqueous humor, iris, and lens) are generally significantly less scattering, so that non-confocal techniques could also be used. Furthermore, since the penetration depth of the light wave fields into the eye is only shallow, the resulting aberrations also pose significantly fewer problems, so that some anterior chamber systems can be used with numerical apertures of over 0.2 and resolutions down to the µm range. The resulting depth of field is between a few micrometers and a few tens of micrometers for various applications.The depth range to be visualized in the anterior chamber is significantly larger and amounts to approximately 6 mm in the tissue (depending on the exact application).

[0004] Therefore, there are basically two optimizations for confocal anterior chamber imaging: a high-resolution confocal scanner that sequentially measures the anterior chamber at many depths of approximately one depth of field each and combines the images. Alternatively, the numerical aperture (nA) can be reduced until a depth of field of approximately 8 mm is achieved in air, which corresponds to an nA = 0.007 at 840 nm and a resulting lateral resolution of approximately 73 µm at 840 nm.

[0005] Non-confocal wide-field OCT systems using fully coherent illumination are also known in the prior art. Since these systems are fully coherent and simultaneously non-confocal, the detected waves can be subsequently propagated / focused into each sample plane, thus partially decoupling the accessible depth range from the lateral resolution. Such methods and coherence tomographs are known from DE 10 2014 115 157 A1. Similar approaches can also be found in DE 10 2014 115 153 A1, DE 10 2014 115 155 A1, DE 10 2015 101 251 A1, and WO 2017 / 137567 A1.

[0006] DE 10 2018 13 0396 A1 describes a method of optical coherence tomography that is illuminated with a scattered, fully coherent illumination wave. Special designs for wide-field image acquisition with a 2D spatially resolved sensor are also presented. The described method is preferably used for volume imaging of biological tissue, such as the human eye. However, it could also be used on technical samples.

[0007] Illuminating a holoscopic system with a scattered, yet fully coherent, illumination wave offers two key advantages. The scattered wave allows for significantly higher illumination intensities without risking damage to the eye. This allows for increased imaging speed, as more light is available. This advantage is less pronounced in anterior chamber systems, where an adjusted illumination wave can influence the proportion of radiation that passes through the iris and reaches the retina. Furthermore, such systems can, in principle, also be illuminated with wavelengths in the range of 1.3 µm to 1.55 µm, which would otherwise not be transmitted to the retina due to the high water absorption of the vitreous body.

[0008] A second advantage of scattered-wave illumination is that it allows multiply scattered stray light components to be distinguished from singly scattered signal light components, thus improving the contrast of images in highly scattering tissues. This advantage, too, is hardly evident in the less scattering tissues of the anterior chamber.

[0009] Depending on the size of the image field to be measured and the lateral resolutions to be achieved, very high pixel resolutions can result that can no longer be achieved with available fast camera sensors. In these cases, the image field is captured in several partial images, which are then calculated / combined to form an overall image. Due to shadowing effects at the boundaries of the partial image fields, the partial images must be captured with a certain overlap, which is determined by the ratio of the depth range to be measured and the depth of field. To achieve a very efficient overall measurement under these circumstances, it is particularly preferable to compose the overall image from as few and as large as possible partial fields.

[0010] Recording the various wavelengths with a 2D camera to calculate the depth function requires a coherent, i.e., phase-locked, image. Residual movements of the eye, when working without a contact lens, as is preferred in this application, limit the time for which this phase correlation can be maintained to approximately 10 to 25 ms. There are prior art methods (e.g., EP 3 517 021 A1) that can estimate movements occurring in retinal images from the partial images, even over time periods of less than one second, and correct them numerically during evaluation. However, anterior chamber images are less spatially structured than retinal images, so these approaches cannot be reliably transferred.

[0011] If you want to measure only the cornea in high resolution, you need wavelength tuning over at least 50 different wavelength images, using suitable illumination with a curved zero-delay surface. If you want to image the entire anterior chamber, you need wavelength tuning over approximately 400 wavelength images, all of which would have to be acquired within approximately 25 ms.

[0012] Very expensive high-speed cameras such as the Photron-SAZ camera, which achieve frame rates up to 20,000 fps at approximately 1 megapixel resolution, are suitable for these frame rate requirements. However, these camera systems are not available with sufficiently good pixel resolutions and are therefore generally unsuitable.

[0013] The invention is therefore based on the object of providing an accelerated, fully coherent wide-field OCT system, with a camera sensor operating at high resolution (i.e., pixel count) but not requiring speeds exceeding 1000 fps. In particular, the system's speed should be suitable for measurements of the anterior chamber of the human eye without a contact lens.

[0014] The invention is defined in the independent claims. The dependent claims relate to preferred developments.

[0015] Wide-field swept-source OCT and the method are used to image a moving object, specifically the anterior chamber of the human eye. The procedure comprises the following steps: Providing wavelength-tuned illumination radiation comprising individual illumination pulses of different centroid wavelengths. The illumination pulses are emitted as a series of illumination pulse pairs, each consisting of a first illumination pulse and a second illumination pulse. In the illumination pulse pairs, the centroid wavelength of the first illumination pulse differs from the centroid wavelength of the second illumination pulse. In several of the illumination pulse pairs, a centroid wavelength of one of the illumination pulses of a previous one of the illumination pulse pairs is repeated. Illuminating the object with the illumination radiation, wherein the illumination radiation is backscattered or reflected in the object as measurement radiation, and imaging the measurement radiation coming from the illuminated object onto a 2D detector. Illumination is preferably carried out using coherent wide-field illumination.Operating the detector according to an image acquisition cycle comprising a sequence of exposure intervals and readout intervals. Synchronizing the illumination pulses and the sequence of exposure intervals and readout intervals such that the illumination pulse pairs are grouped around every second readout interval of the sequence, and for each illumination pulse pair, the first illumination pulse is delivered during a final third of one of the exposure intervals and the second illumination pulse is delivered during a first third of the following exposure interval. Reading image data from the detector for each exposure interval and assigning the image data to the centroid wavelengths of the illumination pulse delivered in the respective exposure interval. Image pairs are generated corresponding to the illumination pulse pairs.Determining changes between the image data of the illumination pulses repeated across the illumination pulse pairs with the same center wavelength and evaluating the image data and using the changes to correct movements of the object in the image data.

[0016] The wide-field swept-source OCT is designed to implement the procedure. It includes: A radiation source configured to emit wavelength-tuned illumination radiation comprising individual illumination pulses of different centroid wavelengths, wherein the illumination pulses are emitted according to the method as a series of illumination pulse pairs with the centroid wavelengths explained above. A 2D detector, also referred to here as a camera, which executes the image acquisition cycle comprising the sequence of exposure intervals and readout intervals. A beam path for illuminating the object with the illumination radiation and for imaging the illuminated object onto the 2D detector. The illumination is preferably carried out as coherent wide-field illumination. A control device that implements the method. It controls the 2D detector and the radiation source and synchronizes them, as mentioned in the method.It generates the aforementioned image pairs, determines changes between the image data of the illumination pulses repeated across the illumination pulse pairs with the same center wavelength, and uses these for referencing when evaluating the image data, thereby correcting movements of the object in the image data.

[0017] The procedure will be explained here using an anterior chamber image, but the description applies to all medical and technical wide-field SS-OCT systems where object movement limits coherent imaging. In particular, the presented solution is also suitable for volume imaging of scattering media for microscopic in vitro applications. Furthermore, aspects described for the procedure apply equally to OCT and vice versa.

[0018] Directly consecutive illumination pulse pairs are spaced apart by a multiple of the separation between the two illumination pulses in the illumination pulse pair. Preferably, the separation between consecutive illumination pulse pairs is at least the duration of a sequence of exposure interval and readout interval, preferably at least 1.5 times, and particularly preferably at least 1.5 times the duration. The separation is defined by the end of the previous illumination pulse and the beginning of the next illumination pulse.

[0019] Each illumination pulse pair is assigned an image pair, hereinafter also referred to as a double image, consisting of two individual images, each of which is assigned the centroid wavelength of the corresponding illumination pulse of the illumination pulse pair. In multiple illumination pulse pairs, a centroid wavelength of one of the illumination pulses of a previous one of the illumination pulse pairs is repeated. This defined structure allows motion correction across the image pairs, since for exactly one individual image of each image pair, at least one other image pair contains an individual image to which the same centroid wavelength is assigned. This allows object movement to be determined for these image pairs by comparing the individual images to which the same centroid wavelength is assigned, and this movement is corrected for the evaluation of the other two individual images of the image pairs.

[0020] This correction is particularly simple if exactly one centroid wavelength of one of the illumination pulses of the immediately preceding illumination pulse pair is repeated in several illumination pulse pairs.

[0021] This, for example, realizes illumination pulse pairs (and thus image pairs) with a wavelength sequence (lambda-0, lambda-1), (lambda-1, lambda-2), (lambda-2, lambda-3), etc. In general, the elements of the sequence of illumination pulse pairs then satisfy the recursive definition (lambda-k, lambda-k+1) with running index k from 1 to n. Of course, the same centroid wavelength can also be repeated as the reference centroid wavelength. This yields the recursive definition (lambda-0, lambda-k) with running index k from 1 to n and lambda-0 as the reference centroid wavelength. Both variants (lambda-0, lambda-k) and (lambda-k, lambda-k+1) are mathematically equivalent. Technically, the former variant is preferable because it requires only one tunable laser.Since the lambda-0 can then also have a larger wavelength separation from all other wavelengths lambda-k, an efficient dichroic merging is also possible in 2-laser systems, which is not possible for the second variant, since the wavelengths sometimes differ only very slightly.

[0022] In principle, any distribution of the centroid wavelengths in the illumination pulse pairs is possible, as long as for each image pair, a correction of an individual image is possible by comparing the other individual image with an individual image of another image pair, thus enabling direct or indirect referencing to another, already motion-corrected or motion-correctable individual image. Thus, a sequence (lambda-0, lambda-1), (lambda-2, lambda-3), (lambda-4, lambda-5), (lambda-6, lambda-7), (lambda-1, lambda-2), (lambda-3, lambda-5), (lambda-4, lambda-7) would be equally suitable.

[0023] Wide-field SS-OCT is preferably implemented as a fully coherent system. These are also referred to as holoscopy systems. The system is particularly preferably implemented with an off-axis detection arrangement, as this allows the suppression of excess noise, the DC component, and most of the autocorrelation signal during evaluation. For this purpose, the light scattered back from the sample is superimposed on the reference wave at an off-axis angle, caused to interfere, and detected. It is important that the camera sensor realizes a 2- to 3-fold higher sampling in the direction of the off-axis angle in order to be able to resolve all spatial frequencies of the interference with the reference wave. A 3-pixel off-axis detection system is particularly preferred. The reference wave exhibits phase shifts of approximately 120° per pixel in the off-axis direction.

[0024] A camera with a pixel resolution well over 1 megapixel is preferred, preferably a global shutter camera. This readout method makes it possible to capture two images in a very short time, which is important for the method described here. However, rolling shutter cameras are also available, which achieve sufficiently fast detection of well over 50 fps with a sufficient number of pixels and can be used for the described detection method.

[0025] For the method, the illumination can be carried out conventionally with a fully coherent spatial single-mode wave or alternatively with a scattered fully coherent multimode wave, as described in DE 10 2018 130 396 A1.

[0026] Microscopic full-field OCT systems, such as those described in C. Apelian et al., Biomedical Optics Express Vol. 7, No. 4, 2016, "Dynamic full-field optical coherence tomography: subcellular metabolic contrast revealed in tissues by interferometric signals temporal analysis," are suitable for very fast volume imaging with microscopic resolution across measurement depths of many depths of field. Since multiple scattering is a significant problem for these microscope systems, scattered wave illumination and analysis that effectively suppresses multiple scattering are particularly preferred for these principles.

[0027] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations indicated, but also in other combinations or on their own, without departing from the scope of the present invention.

[0028] The invention will be explained in more detail below using exemplary embodiments with reference to the accompanying drawings, which also disclose features essential to the invention. These exemplary embodiments are for illustrative purposes only and are not to be interpreted as limiting. To avoid repetition, identical or corresponding elements in different figures are designated by the same reference numerals and are not explained more than once. The figures show: Fig. 1A a timing diagram for a method for OCT imaging, Fig. 1B a block diagram for an OCT device, Fig. 2A-2C possible light sources for the device of the Fig. 1B Fig. 3A-3Dpossible beam geometries for illumination in the device of Fig. 1B .

[0029] For a fast SS-OCT image with an image sensor whose refresh rate, i.e. minimum exposure time, is actually too slow for the desired imaging speed, the Fig. 1 The schematic diagram of the exposure and readout of the detector is used. Fig. 1A A sequence of exposure intervals is shown for a conventional camera, which is designed here as a global shutter sensor, for example. In sequence 2, exposure intervals 4.1, 4.2, etc. follow one another. Each exposure interval represents an integration interval, over which the pixels of the sensor collect electrical charges as long as they are illuminated. The duration of the exposure interval 4.1, 4.2, 4.3, etc. determines the refresh rate, the so-called frame rate. After the individual exposure intervals 4.1, 4.2, 4.3, etc., have expired, the sensor is read out via a readout interval 6.1, 6.2, etc., i.e. the charges that have accumulated in the pixels are transferred to a readout register. Then the next integration, i.e. the next exposure interval, can begin. In detail, this means that the sensor collects charges in its pixels during exposure interval 4.1, and during readout interval 6.1 transfers these to the readout register and cannot accumulate any charges during this time. The next exposure interval 4.2 then begins, in which charges are again collected in the pixels. They are then transferred back to the readout register in the readout interval 6.2, so that the next exposure interval 4.3 can begin. The duration of the exposure intervals is now such that the movement-related problems in the sample, for example, in the eye in the case of an ophthalmic anterior chamber OCT, would no longer be tolerable. Especially with the preferred holoscopy, the required phase rigidity between the individual images would no longer be present during laser tuning.

[0030] Therefore, the exposure, ie the illumination of the sample, is designed in such a way as the process 8 in Fig. 1A illustrated. At the end of the first exposure interval 4.1, a short illumination pulse 10.L0 is irradiated, i.e., only during this time does light fall on the pixels in the exposure interval 4.1. At the beginning of the subsequent exposure interval 4.2, an illumination pulse 10.L1 is emitted, which, according to the swept-source principle of SS-OCT, is at a different wavelength. It is comparatively short compared to the exposure interval 4.2, so that the latter essentially only collects sample light due to the illumination pulse 10.L1. The interval 12 between the two illumination pulses 10.L0 and 10.L1 is illustrated by the double arrow and is essentially limited only by the duration of the readout time 6.1. The distance 12 is so small that no disturbances occur during this time, ie the required phase rigidity between the image collected during the exposure interval 4.1 by the exposure pulse 10.L0 and the image of the exposure interval 4.2, which results from the illumination pulse 10.L1, is given.

[0031] This synchronized sequence is now repeated, with the illumination pulse 10.L0, which is radiated at the end of the exposure interval 4.3, now being followed by an illumination pulse 10.L2, which has a different wavelength than the illumination pulse 10.L.1. The time interval 14 between the illumination pulse 10.L1 and the next pulse is so large that the required phase rigidity would no longer be present here. However, since phase rigidity is again present between the successively radiated illumination pulses 10.L0 and 10.L2, a phase correction can be easily performed across the images of the illumination pulse 10.L1 (recorded during the exposure interval 4.2) and the illumination pulse 10.L2 (recorded during the exposure interval 10.3).

[0032] Thus, the inherently insufficient frame rate of the image sensor is compensated for by creating double images. These double images each correspond to an illumination pulse at a reference wavelength, Fig. 1A the illumination pulse 10.L0, where "L0" symbolizes the reference wavelength lamda-0, and an illumination pulse with wavelength varying according to the wavelength tuning time, symbolized in Fig. 1A by adding "L1" or "L2".

[0033] An S-65A70 camera from Adimec, Eindhoven, Netherlands, is suitable for this application. This global shutter sensor has a resolution of 65 megapixels and can be read at 70 fps. The sensor is set to the longest possible exposure time of 14 ms, which still achieves the full frame rate. After 14 ms, the charges accumulated in the pixels are transferred to a readout register within less than 50 µs (frame overhead time), and the integration of the next frame is started. During the integration of the next frame, the first frame is simultaneously digitized and transferred to the evaluation unit.

[0034] If this scheme is appropriately synchronized with the pulsed illumination for the entire camera, so that, for example, the illumination is only activated for the last millisecond of the first integration interval 4.1 and for the first millisecond of the second integration interval 4.2, two full-resolution camera images are acquired in a total of less than 2.05 ms. No significant motion artifacts occur at such short times.

[0035] Pulsed operation allows measurement speeds to be achieved that are significantly higher than the camera's nominal frame rates. Furthermore, the laser's permissible peak power limits can be further increased.

[0036] In the SS-OCT system, a defined number of images are acquired under illumination with an SS laser at illumination pulses 10.L1, 10.L2, etc., while the laser's k-number (k = 2 pi / lambda) is linearly and continuously tuned. The tuning speed is adjusted to the number of images required for the depth range and to the camera's frame rate.

[0037] For the SS-OCT approach, the laser tuning rate is chosen to be less than or equal to the camera frame rate. The camera frame rate, the irradiation of the illumination pulses, is Fig. 1A adapted to the tuning rate of the camera so that double images are recorded, for example, according to (lambda-0, lambda-1), (lambda-0, lambda-2), (lambda-0, lambda-3), ... (lambda-0, lambda-n). Due to the short recording time, each double image is fully coherent; however, the double images may exhibit motion artifacts due to the distance 14, which does not fulfill the condition of phase rigidity, which is essential for the evaluation. However, since an image is repeatedly recorded with the reference wavelength lambda-0 for each double image, the phase position within the double images is fixed and can therefore also be indirectly coordinated across the double images. For example, in the i-th double image, the difference phase between the images at lambda-0 and lambda-i is determined and used for further evaluation.This compensates for the phase errors caused by motion artifacts and the images for lambda-0 to lambda-i are again coherent with each other with regard to the evaluation.

[0038] This applies to axial motion artifacts. Lateral motion artifacts can also be optionally corrected by cross-correlating the reference images with lambda-0. The result is the displacement vector between the images, which can be used to compensate for the displacement of the lambda-0 images and the corresponding lambda-i images.

[0039] Of course, the phase referencing does not have to be done with the wavelength lambda-0, but can be done with any wavelength from the wavelength sequence or even with an additional wavelength, which is provided, for example, by a fixed frequency reference laser.

[0040] If the camera achieves high frame rates, it may be sufficient to perform a reference measurement at the fixed wavelength only every 3, 10, or more wavelength images. Generally, the reference measurement is performed every m images (m is a natural number greater than zero).

[0041] The intervals between reference images of the same frequency can also be chosen so large that measurable motion artifacts / phase shifts occur. In this case, it is particularly preferable to unwrap the phase positions of the reference images by 2 pi, interpolate the phase changes for all intervals, and use these to correct the different wavelength images.

[0042] Fig. 1B shows a schematic of an OCT 12 that acquires three-dimensional images of an eye 14, specifically of the anterior chamber 16. The described embodiment is based on the example of a fiber-based swept-source system, but can also be applied to free-beam setups. Source radiation from a wavelength-tunable radiation source 18, for example, a corresponding laser, as will be explained below using the example of Figs. 2-4, is coupled from an output 19 of the radiation source 18 into a fiber 20. The source radiation lies, for example, in the infrared wavelength range, whereby this wavelength range is also referred to as "light" in this description. This term encompasses all radiation of the electromagnetic spectrum that satisfies the laws of optics. The eye 14 is imaged by optics (not further designated) onto a detector 22, which detects the Fig. 1A explained reading scheme according to process 2.

[0043] According to the fiber-based design, the fiber 20 terminates in a splitter 21, which splits the source radiation into a measuring arm 24 and a reference arm 26. A fiber 28 connects to the splitter 21 in the measuring arm 24, and the illumination radiation B exiting at the fiber end is modified with respect to the illumination modes by means of an illumination optics 30 – in particular, a diffusion disk 40 is provided for this purpose, the effect of which is described, for example, in DE 10 2018 130 396 A1 – and then guided to a beam splitter 32. From there, it reaches the anterior chamber 16 via a front optics (not shown).

[0044] This illumination radiation is backscattered in the anterior chamber 16 from various depths z within a depth of field range. The backscattered radiation is collected as measuring radiation M and reaches the detector 22. The beam splitter 32 thus separates the measuring radiation M from the illumination radiation B. The detector 22 has spatial resolution, i.e., it allows resolution of the intensity distribution across the beam cross-section according to the holoscopy principle. The detector 22 is preferably conjugated to the eye pupil. However, since the light wave field, once it is in a plane with known magnitude and phase, can be converted to any other plane by numerical propagation, detection can also take place in other planes, e.g., to a plane in the anterior chamber 16.

[0045] The radiation separated by the splitter 21 into the reference beam path 26 enters a fiber 34 and is radiated obliquely onto the detector 22 via a deflection mirror 36 and another fiber 38 with an adjustable path length. This results in a superposition of the reference radiation and the measurement radiation M, here by means of so-called off-axis detection. The path length adjustment 37 is designed as a free beam path in the illustrated embodiment. This is optional, as is the use of the mirror 36. Various measures for adjusting the optical path length of a beam are known in the prior art.

[0046] To perform the image generation, the OCT 12 has a control unit C which controls the wavelength tuning of the radiation source 18 and the operation of the detector 22 according to the Fig. 1A The system synchronizes the phase relationship from the double images to compensate for the distance 14 and generates corresponding holoscopic images. The latter is known in the prior art.

[0047] For fundamental physical reasons, the laser cannot be modulated during the measurement to achieve the degree of temporal coherence required for the measurement. Therefore, the emitted light from the laser is amplified, if necessary, using an amplifier (e.g., an SOA) and then switched by a fast optical switch. If an SOA is used, it can be incorporated into the optical circuit.

[0048] For the described method and device, swept-source laser sources 18 are preferred, whose wavelength characteristics are so predictable that pulse control can be achieved without online monitoring of the laser wavelength, i.e., without a wavelength measurement arrangement (k-clock). If this is not possible, a k-clock arrangement is used, as is known in the art for controlling the synchronization of optical switches.

[0049] J. Kühn et al., Optics Express Vol. 15, No. 12, 2007, "Real-time dual-wavelength digital holographic microscopy with a single hologram acquisition," describes a digital multi-frequency holography method in which two wavelengths, lambda-0 and lambda-n, are recorded simultaneously in a single camera image. The reference waves of the two wavelengths are irradiated from different directions so that they can be separated from each other during evaluation by suitable lateral filtering. The advantage of this illumination arrangement is that the reference image is captured at exactly the same time as the measurement image, thereby suppressing motion artifacts even better. This can now be achieved with the rolling shutter camera, and the rapid switching of the lasers for the realization of the integration time is achieved through camera integration. This makes it easier to implement the laser sources.A disadvantage of this arrangement is that the image must be sampled twice as densely in the direction perpendicular to the off-axis detection to achieve filtering. This effectively results in half the camera pixels being lost. Furthermore, the two reference waves reduce the SNR by 3 dB. This configuration is therefore preferred for measuring very dynamic objects.

[0050] If the described method is combined with scattered-wave illumination, the evaluation is slightly modified. In this case, it is not sufficient to subtract the phase of the reference measurement, since the propagation of the scattered wave into the sample creates phase and amplitude changes that are not spatially variable and thus change when the object is moved relative to the measuring device. The spatial frequencies introduced by the scattered speckle illumination mix with the unlimited spatial frequency spectrum of the object to be measured. This mixed frequency spectrum is filtered by the subsequent filtering, which is implemented using a detection aperture. A clear "unmixing" of the transmitted, filtered, and detected spatial frequency spectrum is then no longer possible from a single wavelength image stack.Therefore, for the further discussion, a distinction is made between two evaluation modes: a simple mode in which exactly one wavelength image stack is evaluated and an extended evaluation mode in which several, preferably about 10 to 40 wavelength image stacks are coherently calculated.

[0051] In the simple embodiment, no attempt is made during the evaluation to separate the sample spatial frequency spectrum, which has been mixed with the scattered illumination angle spectrum. Instead, the final image displays the mixture of detection speckles, which arise from backscattering in the object, with the illumination speckles. To ensure that this reconstruction method does not cause artifacts, it is important that the illumination speckles differ from each other and from the object laterally and axially by significantly less than one speckle grain during the acquisition of all images that are to be coherently calculated. For simple evaluation, the scattering disk is therefore imaged in the central object plane so that the speckle patterns of the different wavelengths are approximately identical, except for a wavelength-dependent propagation phase characteristic of the method.The achievable depth range scales with lambda 2< / (nA 2< *delta-lambda).

[0052] In addition, the images should be acquired as quickly as possible. If the images only shift around a fixed point, as is the case with images of a human retina fixed to a fixation light, an alternative is to choose the illumination aperture so small that the resulting illumination speckles are larger than the shift amplitude.

[0053] A third option is to choose an illumination aperture significantly smaller (less than three times) than the detection aperture. In this case, only minor artifacts arise, manifesting as higher, but still acceptable, image noise. However, in this case, the illumination wave is usually so severely restricted in its etendue that the usable light thresholds are only slightly higher than those of single-mode illumination.

[0054] If fast motion artifacts with high amplitude occur and high demands are placed on the lateral resolution and image quality to be achieved, or if very deep object structures are to be measured that extend beyond the depth range described in the previous section, an extended analysis is preferred. Several independent wavelength image stacks are acquired and coherently processed. There are three variants of this extended analysis, each with its own distinct properties. Some of these variants can also double the lateral resolution and suppress multiple scattering in the images.

[0055] For the extended evaluation described below, it is preferable to image the diffusion disk 39 into the object plane; however, for some variants, it is also possible to position the diffusion disk, for example, close to the pupil. This is particularly important in lensless embodiments, as known in the prior art. The specific problem with this type of evaluation is that all images must be coherently processed. By capturing multiple image stacks, even with cameras with high lateral resolution, a phase-locked image acquisition is realized, as will be explained, even over several seconds.

[0056] Fig. 2 shows a light source 19 with only one laser 42 for the following first and second illumination variants.

[0057] The first illumination variant has already been mentioned. There is only one tuned laser source 42, with tuning rates on the order of the frame rate of camera 22. The fast switching 40 of source 19, synchronized with the recording of camera 22, appropriately cuts out and detects the double images with the wavelengths (lambda-0, lambda-n). In this case, only one source 42, an optional splitter 44 for an optional k-clock 48 (see above), and a fast switch 46 are required. The disadvantage of this design is that the time interval between lambda-0 and lambda-n cannot be chosen arbitrarily short, as it is limited by the tuning time of laser 42. Furthermore, with this type of detection, the integration time for a single image lambda-i is limited, as laser 42 is continuously tuned and only a certain wavelength drift can be tolerated during the integration time.

[0058] The second illumination variant differs from the first only in its laser control. In this case, the laser 42 is not continuously tuned, but switches between fixed wavelengths. At each new wavelength, the laser requires a specific settling time of approximately 1 to 100 ms to return to stable long-coherence operation. In this case, the integration interval 4.1 etc. can be chosen significantly longer, so that it is only limited by the motion artifacts. However, the minimum time interval within the double image corresponds to the settling time. This solution is therefore particularly preferred for laser systems that settle into stable operation in less than approximately 10 ms. During the settling time, the laser is blocked by switch 46 to avoid corrupting the camera signals.

[0059] A third lighting variant according to Fig. 3 operates with two lasers 42, 50, a fixed frequency laser 50 with the wavelength lambda-0 and a second tunable laser 42. The tunable laser 42 is operated as in Fig. 2either switched between discrete wavelengths lambda-i or slowly tuned continuously. Slow means, for example, in this case that if 400 wavelengths lambda-i are used, the laser 42 is tuned once in approximately 11 seconds in order to be able to record 400 double images lambda-0, lambda-i with the camera during this time. In this case, a fast optical switch 52 combines the two laser beams and also provides fast optical switching. This variant is technically the most complex, but allows very fast double images and permits long integration intervals. In this constellation, it is particularly preferred if an SOA / optical post-amplifier 54 is implemented behind the switch 52 in order to be able to amplify both lasers 42, 50. This variant can also preferably be combined with the described simultaneous 2-reference recording.In this case, an intensity splitter is used instead of the optical switch 52, and exposure is achieved through the camera's integration interval. The reference waves for the k-clock 48 are split before being combined in the splitter 44a, 44b. This limits the additional effort to the implementation of the second fixed-frequency laser 50. However, since most OCT images only require coherence lengths in the 10 mm range, technically simple reference laser diodes can be used.

[0060] If the described imaging technique is used for images of the cornea and the anterior chamber 16 of the human eye 14, the number of images to be acquired for specific applications can be minimized by using specific beam geometries. Various beam geometries and their advantages and disadvantages are described below.

[0061] The advantage of the OCT system described here is microscopic resolution in the micrometer range over tissue depths of a few millimeters. Parameterization allows the system's depth of field and the depth resolution achieved by coherence measurement to be adjusted independently of each other. However, for fully coherent 3D reconstruction, it is important that the achieved coherence depth resolution be selected to be better than or equal to the system's depth of field.

[0062] If microscopic resolution is to be achieved only in corneal tissue 60, the illumination radiation should preferably be directed in such a way that a curved zero-delay surface is created near the anterior surface of the cornea. The shape and position of the zero-delay surface are equally influenced by the illumination and backscattered detection waves. To create a zero-delay surface on the anterior surface of the cornea, the Fermat principle can be applied. If the anterior surface of the cornea is imagined to be mirrored and the illumination is implemented in such a way that it is imaged into the detection by corneal reflection, the zero-delay is adjusted to the corneal surface by adjusting the length of the reference arm. Three geometries are conceivable for this, as shown in Fig. 4A-C. However, mixed forms can of course also be realized. Detection takes place with a specific numerical aperture. In Fig. 4A-C, however, only the main rays are shown in a simplified manner.The figure shows beam geometries for zero delay on the anterior corneal surface 61 (illumination B shown in dashed lines, detection M shown in crossed lines). The mean radius of curvature of the cornea 60 is 8 mm. In Fig. 4A, the detection is aimed approximately 4 mm behind the corneal vertex 62. In Fig. 4B, the illumination is aimed approximately 4 mm behind the corneal vertex 62. In Fig. 4C, both are aimed approximately 8 mm behind the corneal vertex 62.

[0063] The advantage of one of the beamlines shown in Fig. 4 is that, with a measurable depth range of approximately 700 µm in tissue or approximately 1 mm in air, the entire cornea can be imaged with high resolution. The disadvantage of this arrangement is that the very bright corneal anterior reflex can overwhelm the dynamic range of the image. Furthermore, larger optic diameters are required to achieve the beamline.

[0064] Fig. 5 shows an improved illumination system with a parallel beam and telecentric detection. This is better suited for imaging the entire anterior chamber and for imaging smaller subfields. Furthermore, for imaging of Schlemm's canal, for example, the angle of incidence can be selected more flexibly to improve imaging by minimizing the path through the sclera. For this purpose, it is particularly preferred if the measuring device, in addition to x, y, and z adjustment via a mechanical stage used in ophthalmic devices, also has a swivel device (possibly a tilt device), as is common in fundus cameras and slit lamps.

[0065] It is preferable to arrange the optical switch before the separation of the signal illumination wave and the reference wave in order to also switch the reference wave and thus minimize the DC component in the camera images.

[0066] The methods developed in the state of the art for motion compensation attempt to compensate for movements only within a wavelength image stack. To do this, an attempt is made to distinguish between signal changes caused by the change in wavelength and signal changes caused by motion artifacts during a wavelength scan, and then to compensate for the motion artifacts. This has three application limitations that make it impossible to transfer it to the problem described here. The method is only suitable for correcting minor movements within the image stack. This allows the coherence time of a few milliseconds to be extended by a factor of about 10. To achieve the extremely fast camera frame rates required for this, high-speed cameras must be used, which have a reduced resolution of only about 1 megapixel.In addition, the objects to be recorded must contain sufficient lateral and axial structures to allow estimation of motion artifacts. This is particularly difficult due to the lower pixel count compared to the high-resolution camera sensors used here. In summary, prior art methods estimate motion from fast wavelength image sequences, whereas the method and device presented here also estimate motion from individual images.

[0067] Due to the motion artifacts that occur both within the acquisition of the image stacks and between the image stacks, the images of the same wavelength in the different stacks have a different random shift of the illumination speckle pattern with respect to the object to be measured, which should be larger than a speckle grain.

[0068] In the following, the particularly preferred first sub-variant of the extended evaluation will be described first and then only the differences between the other two possible sub-variants will be discussed.

[0069] For the extended evaluation, in a first step the movements between the image pairs are estimated by comparing the recording with the reference wavelength and thus the recording of the second wavelength, which was recorded promptly without movements, is digitally corrected so that all recordings of the multiple image stacks are subsequently coherent with each other.

[0070] In the second step, the images from the various stacks are rearranged, and a higher-resolution image is reconstructed from all images of the same wavelength. This image also exhibits suppressed multiple scattering, depending on the number of stacks processed. To do this, the calculated signal field strengths of the various images are simply added together. The phase correction in the first step constructively superimposes the singly scattered signal components, while the multiply scattered signals are superimposed in a phase-uncorrelated manner, thus partially averaging out the signals.

[0071] The third step is again identical to the familiar holoscopic reconstruction; the resulting enhanced-resolution images of the various wavelengths are combined using a Fourier transform to create a depth-resolved final image. More complex 3D analysis methods also exist in the state of the art that can be used alternatively, but are referred to simply as Fourier transforms.

[0072] For the particularly important first step of the evaluation, the reference wavelength image is extracted from each acquired image pair and compared with a fixed comparison image, e.g., the first reference wavelength image. The two images are laterally cross-correlated, and then the lateral shift is corrected. The same shift is also corrected in the corresponding second wavelength image. In the next step, the field strengths of the first and nth reference images are compared. The phase values ​​will differ for three physically different reasons.

[0073] Firstly, because the two reference images may have changed in absolute phase due to axial movements. These changes can lead to a constant global phase, but in most applications, higher lateral orders such as tilt angles (phase ramps) and defocus (quadratic lateral phase distributions) will also occur and must be corrected. However, these phase functions are locally highly correlated, so the phase change does not need to be estimated individually for each pixel; instead, a parameterized phase function can be fitted with a few to tens of parameters.

[0074] Second, the scattered illumination wave with its typical speckle pattern will result in a random initial phase in each speckle grain. The excitation field strength will also vary laterally and axially across the object due to these speckles. These speckle amplitudes and phases are assumed to be known. They can be derived from the design of the diffuser or measured in a plane using a calibration measurement with a mirror as the object. From there, they can be transferred / converted to all points on the illuminated object using digital propagation methods. The mirror / concave mirror is designed to project the exit pupil of the scattered illumination into the entrance pupil of the detection optics. If the beam path is telecentric on the object side, plane mirrors are used; otherwise, concave mirrors are used.The measured field distribution of the scattered excitation wave is referred to below as the calibration measurement and is recorded individually for each wavelength. The measured signal field strength image is divided into a complex pattern by the known excitation field strength / calibration measurement to compensate for this influence of speckle illumination. Both quantities are determined and calculated in the conjugate image plane / zero delay. Since this division can result in double spatial frequencies, the signal field strength image and calibration image are interpolated to double the pixel density prior to this operation.

[0075] The third effect arises from an interaction of the illumination spatial frequencies with the spatial frequencies of the object's backscatter coefficient. Thus, the amplitude and phase values ​​at an object point change depending on the gradient of the excitation field strength at that point, which will be different for each position of the illumination speckle pattern relative to the sample.

[0076] If, for example, the object is illuminated with an illumination speckle pattern that has the same angular aperture as the detection, the structured speckle illumination makes twice the spatial frequency values ​​of the object accessible than would be calculated from diffraction-limited detection alone. However, the detection only transmits the single spatial frequency spectrum. For this reason, in this example, only a quarter of the spatial frequency information is transmitted in one image, and it is fundamentally impossible to determine this phase and amplitude effect from a single image. When a sufficient number of images of the same wavelength but with different illumination speckle shifts are later combined, all spatial frequencies up to twice the detection limit frequency can be unambiguously reconstructed. However, this requires coherent combination, for which the shift phases must have been determined and corrected beforehand.Therefore, this effect cannot be compensated for when determining the phase shift of the individual images, and its effects must be dealt with. After correcting the detected signal field strength images by the excitation field strengths, the effect described in this example leads to a phase decorrelation of approximately 75%. To still be able to estimate the phase position of the reference images relative to each other, averaging must be performed laterally across several (many) pixels. The global phase can be determined by averaging all pixel phases. For practical applications, however, the described approach, which involves fitting a parameterized phase function with significantly fewer parameters than detection pixels, is by far the preferred approach.Once the phases have been determined in this way, the phases are converted from the reference image wavelength Phi-ref to that of the measurement image wavelength Phi-mess (Phi-mess = Phi-ref * lambda-mess / lambda-ref) and corrected in these measurement images with variable wavelength.

[0077] The last two steps of the evaluation—the coherent summation of the field strengths of all images of the same wavelength after the described correction and the subsequent Fourier transformation across the different wavelengths—can also be performed in reverse due to the linearity of the Fourier transformation. For this purpose, each corrected image stack is Fourier transformed individually, and then the complex resulting fields are coherently added.

[0078] The first sub-variant of the extended analysis has the following properties: The diffusion disk does not necessarily have to be imaged into the object plane, making this variant also suitable for lensless systems. However, the excitation speckle distribution must be known from the calibration measurement. Depending on the choice of illumination aperture, the resulting image will have twice the lateral resolution. Depending on the number of reconstructed image stacks, multiple scattering in the resulting image will be suppressed because it is partially averaged out. However, the resulting images will be fully speckled. To partially despeckle the images, four pixels in the resulting image can be incoherently averaged, but this reduces the lateral resolution to the normal detection resolution.

[0079] In the second sub-variant of the extended evaluation, the signal images are first divided by the calibration measurement, then the lateral displacements and the axial phases are compensated for motion artifacts. However, the Fourier transform over the wavelengths is then calculated for each image stack. The resulting depth-resolved images of the different image stacks are then added incoherently, i.e. the absolute values ​​of the function are calculated and added. The result is a final image with normal lateral resolution limited by detection diffraction, in which multiple scattering is not suppressed, but which can also be completely despeckled depending on the number of stacks processed. This despeckle removal applies to both the singly scattered and the multiply scattered signal components, so that the final images will appear very smooth / low-noise.By increasing the image contrast, it is then possible to virtually suppress the multiple scattering visible in the image. For this variant, the diffusion disk does not need to be imaged into the sample, making it suitable for lensless systems.

[0080] In the third sub-variant of the extended evaluation, the signal image is not divided by the calibration measurement. However, the lateral displacements and the axial phase changes occurring due to movement are corrected as described in the other sub-variants. As with the first sub-variant, the images are first added coherently and then the Fourier transform is calculated over the wavelengths, or in the reverse order. This evaluation variant only works if, as with the simple evaluation, the scattering disk is sharply imaged into the sample and the displayed maximum depth range is maintained. In this case, the initial phases of the scattered excitation wave in the zero delay are unknown, but they are approximately the same at all wavelengths and therefore cancel out during evaluation.Therefore, in this variant, the singly scattered signal components are superimposed constructively and the multiply scattered signal components are superimposed uncorrelatedly, resulting in images that exhibit suppressed multiple scattering. However, since the intensity fluctuations of the scattered illumination wave, which also influence the transmission of spatial frequencies, are not corrected, the higher diffraction orders will be emphasized, and the resulting image will only exhibit the simple resolution limited by detection. However, the resulting images are at least partially despeckle-free. The evaluation according to the third sub-variant is also described in principle in DE 10 2018 130 396 A1.

[0081] A feature of the described arrangement is the very high degree of parallelization with multi-megapixel resolution. This results in very low excitation intensity per pixel, requiring either very long integration times or the use of very powerful lasers. Therefore, the following describes specific light source system designs that are preferred for this application.

[0082] The illustrated arrangements only show the aspects of the absolute beam paths traversed, not the imaging performance or image field positions. Correcting such large image fields of up to approximately 12 mm for numerical apertures of 0.2 and larger, while simultaneously maintaining diffraction-limited resolution, is optically very difficult to achieve, especially for convexly curved image field planes. A major advantage of fully coherent wide-field OCT, however, is that known image fields can be extracted from the data during evaluation by numerically propagating the detected light wave fields to the pupil, correcting the phase function of the aberrations there, and then numerically propagating the wave back to the field plane. This procedure is well known in the art.In this case, it is particularly preferred that the optics are corrected for maximum sharpness in the edges of the image field at the expense of correction in the center of the image field (especially in the case of field curvature) in order to minimize the edge effects / overlap area in the numerical propagation.

Claims

1. A wide-field swept-source OCT method for imaging a moving object (14), in particular the anterior chamber (16) of the human eye, wherein the method comprises the following steps: - providing illumination radiation (B), which is tuned in the wavelength and comprises individual illumination pulses (10.L0, 10.L1) of different centroid wavelength, wherein the illumination pulses (10.L0, 10,L1, 10.L2) are provided as a series of illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2), each consisting of a first illumination pulse (10.L0) and a second illumination pulse (10.L1, 10.L2), wherein in the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) the centroid wavelength of the first illumination pulse (10.L0) differs from the centroid wavelength of the second illumination pulse (10L1, 10.L2), and in a plurality of the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) at least one centroid wavelength of one of the illumination pulses (10.L0) of a preceding one of the illumination pulse pairs (10.L0, 10.L1; 10.L0, 10.L2) is repeated, - illuminating the object (14) with the illumination radiation (B) and imaging the illuminated object (14) onto a 2D detector (22), - operating the detector (22) according to an image recording cycle comprising a sequence (2) of exposure intervals (4.1, 4.2, 4.3, 4.4) and readout intervals (6.1, 6.2, 6.3), - synchronizing the emission of the illumination pulses (10.L0, 10,L1, 10.L.2) and the operation of the detector (22) in such a way that the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) are grouped around every second one (6.1, 6.3) of the readout intervals (6.1, 6.2, 6.3) of the sequence (2) and, for each illumination pulse pair (10.L0, 10L1; 1 0.L0, 1 0.L2), the first illumination pulse (10.L0) is emitted during a last third of one of the exposure intervals (4.1, 4.2, 4.3, 4.4) and the second illumination pulse (10L1; 10.L2) is emitted during a first third of the next of the exposure intervals (4.1, 4.2, 4.3, 4.4), and - reading image data of each exposure interval (4.1, 4.2, 4.3, 4.4) of the detector (22) and assigning the image data to the centroid wavelengths of the illumination pulse (10.L0, 10,L1, 10.L.2) emitted in the respective exposure interval, wherein image pairs consisting of single images are generated according to the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2), - determining changes in the image pairs between the single images to which the same centroid wavelength is assigned, and - evaluating the image data and using the changes to correct movements of the object (14) in the image data.

2. The method as claimed in claim 1, wherein in a plurality of illumination pulse pairs (10.L0, 10L1 ; 10.L0, 10.L2) exactly one centroid wavelength of one of the illumination pulses (10.L0) of the immediately preceding one of the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) is repeated.

3. The method as claimed in claim 2, wherein the same centroid wavelength is repeated as the reference centroid wavelength.

4. The method as claimed in any of the above claims, wherein a temporal distance (12) of the illumination pulses of the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) is selected such that the single images of each image pair differ in location from the object (14) by less than one speckle grain of the illumination.

5. The method as claimed in any of the above claims, wherein a movement during the imaging of the object is realized by a motorically moved object carrier.

6. The method as claimed in any of the above claims, wherein a non-living in-vitro object is imaged and illuminated for this purpose with illumination radiation (B) formed as scattering wave illumination, wherein a plurality of wavelength image stacks are recorded and a displacement between illumination speckles and the object is achieved by a motorized movement of the object.

7. The method as claimed in any of the above claims, wherein this motorized movement of the object is carried out only between the image stacks and not within the image stacks.

8. The method as claimed in any of the above claims, wherein the measurement radiation is superimposed with reference radiation on the detector (22), wherein different reference radiation directions are provided for the single images of each image pair, and the two single images are separated in an image evaluation based on the reference radiation directions.

9. A wide-field swept-source OCT for imaging a moving object (14), in particular the anterior chamber (16) of the human eye, wherein the OCT (12) comprises: - a radiation source (18), which emits illumination radiation (B) which is tuned in the wavelength and comprises individual illumination pulses (10.L0, 10.L1) of different centroid wavelength, wherein the illumination pulses (10.L0, 10,L1, 10.L2) are emitted as a series of illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2), each consisting of a first illumination pulse (10.L0) and a second illumination pulse (10.L1, 10.L2), wherein in the illumination pulse pairs (10.L0, 10L1 ; 10.L0, 10.L2) the centroid wavelength of the first illumination pulse (10.L0) differs from the centroid wavelength of the second illumination pulse (10L1, 10.L2), and in a plurality of the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) at least one centroid wavelength of one of the illumination pulses (10.L0) of a preceding one of the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) is repeated, - a 2D detector (22), which performs an image recording cycle comprising a sequence (2) of exposure intervals (4.1, 4.2, 4.3, 4.4) and readout intervals (6.1, 6.2, 6.3), - a beam path (24) for illuminating the object (14) with the illumination radiation (B) and for imaging the illuminated object (14) onto the 2D detector (22), and - a control device (C), which controls the 2D detector (22) and the radiation source (18) and is configured to synchronize the radiation source (18) and the 2D detector (22) in such a way that the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2) are grouped around every second one (6.1, 6.3) of the readout intervals (6.1, 6.2, 6.3) of the sequence (2) and, for each illumination pulse pair (10.L0, 10L1; 10.L0, 10.L2), the first illumination pulse (10.L0) is emitted during a last third of one of the exposure intervals (4.1, 4.2, 4.3, 4.4) and the second illumination pulse (10L1; 10.L2) is emitted during a first third of the next of the exposure intervals (4.1, 4.2, 4.3, 4.4), - wherein the control device (C) is further configured -- for reading image data for each exposure interval (4.1, 4.2, 4.3, 4.4) of the detector (22) and for assigning the image data to the centroid wavelengths of the illumination pulse (10.L0, 10,L1, 10.L.2) emitted in the respective exposure interval, and for generating image pairs according to the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2), -- for determining changes between the image data of the illumination pulses (10.L0) with the same centroid wavelength repeated across the illumination pulse pairs (10.L0, 10L1; 10.L0, 10.L2), and -- for evaluating the image data and for using the changes to correct movements of the object (14) in the image data.

10. The OCT as claimed in claim 9, wherein the radiation source (18) comprises a swept-source laser (42), which has a tuning repetition rate which is not lower than a frame rate of the detector (22) defined by the duration of the exposure interval (4.1, 4.2, 4.3, 4.4) and readout interval (6.1, 6.2, 6.3).

11. The OCT as claimed in claim 9 or 10, wherein the radiation source (18) comprises an optical switch (46) or switch (52) controlled by the control device (C) for synchronization.

12. The OCT as claimed in any of claims 9 to 11, wherein a coherence depth resolution is better than or equal to a depth of field defined by the beam path.