Imaging apparatus and method for in vivo full-field interferometric imaging of a sample - Patents.com

Abstract

The invention relates to an imaging device (2) for in vivo full-field interferometric imaging of a sample, the imaging device comprising: a primary light source (14) configured to output a probe beam (20) having low spatial coherence and low temporal coherence length; a sample arm (26); a reference arm (28) consisting of a reference reflective surface (30); a primary interferometer (16) comprising at least a primary beam splitter (24); a camera (18) provided by the primary interferometer (16) and configured to acquire at least one interference image representative of a sample placed in the sample arm (26); a sample beam splitter (19) arranged in the sample arm; and a sample position detection system (10) arranged outside the sample arm and having a detection axis intersecting the sample beam splitter (19), the sample position detection system being configured to output a position signal representative of the position of the sample along the detection axis.

Description

[Technical Field]

[0001] The present invention relates to an imaging apparatus for in vivo full-field coherent imaging of scattering samples.

[0002] The present invention relates to the field of in vivo imaging, and more particularly to in vivo imaging of ophthalmic tissues. [Background technology]

[0003] Optical coherence tomography (OCT) is an interferometric technique that has become a powerful imaging modality. OCT has a wide range of applications, particularly in biomedical fields such as ophthalmology, dermatology, cardiovascular medicine, and gastroenterology. Examples of optical coherence tomography include full-field OCT.

[0004] For example, document WO2020 / 065068A1 discloses an imaging device for full-field OCT designed for in vivo imaging of ophthalmic tissues.

[0005] The imaging device includes a full-field optical coherence tomography (FF-OCT) system. Furthermore, the imaging technique implemented in this device is based on the use of light backscattered by a sample (i.e., in the case of a biological sample, the microscopic cellular and tissue structures of the sample) when illuminated by a probe beam emitted by a light source of the FF-OCT system, which light source has a low spatial and temporal coherence length. The sample is typically the subject's eye.

[0006] A microscope objective is placed in front of the specimen to focus the light from the light source onto the specimen, specifically the cornea. Alternatively, for retinal imaging, the microscope objective is removed and the cornea and lens of the eye ensure focusing of the probe beam onto the retina.

[0007] The FF-OCT system further includes an interferometer including a reference arm and a sample arm separated by a beam splitter. In this case, the sample is placed in the sample arm of the interferometer. The presence of the interferometer allows for the generation of an interference light signal representative of light selectively emanating from a given slice of the sample volume while rejecting light emanating from the remainder of the sample. This is because the low temporal coherence of the light source allows for the separation of backscattered light at a virtual slice of depth within the sample.

[0008] The FF-OCT system further includes a camera that captures two-dimensional interference images representative of the interference light signal over time.

[0009] Next, at least one FF-OCT image can be obtained by processing the interference image obtained based on the optical interference signal, wherein the FF-OCT image is a front image of the sample, i.e., an image of the sample in a plane perpendicular to the optical axis of the sample arm.

[0010] Additionally, a spectral-domain optical coherence tomography (SD-OCT) system is provided for tracking the position of the eye. The SD-OCT system includes a secondary light source, the secondary beam of which propagates along the probe beam and is coupled through the illumination arm of the FF-OCT device to impinge on the sample. Analysis of photons backscattered by the sample in response to illumination by the secondary beam allows the position of the sample to be determined in real time. Knowing the position of the eye allows for real-time correction of eye movements, particularly the direction of propagation of the probe beam.

[0011] Furthermore, by scanning the secondary beam using a scanner, it is possible to view the cross section of the sample.

[0012] However, such imaging devices are not entirely satisfactory.

[0013] In fact, the FF-OCT and SD-OCT systems are highly interrelated due to the shared optical elements in such imaging devices, making assembly and alignment of such devices extremely challenging.

[0014] Furthermore, in such known architectures, the scanner of the SD-OCT system is located far from the microscope objective (typically 30 cm). As a result, only light propagating at a small angle relative to the propagation direction of the probe beam enters the entrance pupil of the microscope objective. As a result, the cross-sectional image provided by the SD-OCT system contains only a few hundred pixels in the lateral direction (i.e., perpendicular to the propagation direction of the probe beam in the sample arm of the interferometer). This results in the SD-OCT system providing an SD-OCT image of unsatisfactory quality. Furthermore, such SD-OCT systems have a small field of view (approximately 1 mm).

[0015] Furthermore, in this configuration, the scanner is not optically conjugated to the back focal plane of the objective. As a result, the secondary beam is not perpendicular to the sample surface, resulting in low-signal SD-OCT images. As a result, the signal-to-noise ratio in the cross-section is too low to allow real-time assessment of the health of the ocular layers.

[0016] Furthermore, in such known devices, it is highly impractical to achieve conjugation of the scanner to the back focal plane of the objective. In fact, even if such conjugation could be achieved by adding a 4F setup to the SD-OCT system, the long path of the SD-OCT beam in the FF-OCT system (approximately 20 cm) would make the 4F setup unacceptably long (approximately 80 cm), making it unsuitable for commercial use.

[0017] Furthermore, because the SD-OCT system is coupled via the illumination arm of the FF-OCT device, the beam splitter located in the interferometer of the FF-OCT device causes a 75% signal loss on the SD-OCT system side compared to when the secondary beam output by the SD-OCT system is directly incident on the sample, which is detrimental to the calculation of high-quality cross-sectional images of the sample and the accurate positioning of the moving sample.

[0018] SUMMARY OF THE INVENTION An object of the present invention is to provide an image forming apparatus that overcomes such problems. Summary of the Invention

[0019] To this end, the invention relates to an imaging device of the aforementioned type, comprising: a full-field OCT imaging system for providing at least one interferogram representative of a front-on image of the sample, the full-field OCT imaging system comprising: - a primary light source configured to output a probe beam having low spatial coherence and a low temporal coherence length; a first-order interferometer, a sample arm for receiving the sample; a reference arm with a reference reflective surface; A primary beam splitter for separating the sample arm and the reference arm. A first order interferometer comprising at least: a camera configured to acquire at least one interferometric image, Each interference image represents a first order optical interference signal resulting from interference occurring at each point in the imaging field when the sample is placed in the sample arm, and the interference is a reference wave obtained by reflection of an incident light wave of the probe beam on a primary surface of the reference reflecting surface conjugate to the point of the imaging field; a sample wave obtained by scattering an incident light wave of the probe beam in an elementary volume of a slice of the sample at a predetermined depth, the elementary volume being conjugate to the point in the imaging field; and Cameras are what happens between; and The imaging device further comprises: - a sample beam splitter arranged in the sample arm downstream of the primary beam splitter in a direction from the primary light source towards the primary beam splitter; - a sample positioning system disposed outside the sample arm, the sample positioning system having a respective detection axis (Δ) that intersects with the sample beam splitter, the sample positioning system configured to output a position signal representative of a position of the sample along the detection axis (Δ). Equipped with.

[0020] In fact, such a structure allows the full-field OCT imaging system and the sample position detection system to share minimal components (i.e., the sample beam splitter), making these two systems nearly optically independent, which simplifies manufacturing, assembly, and maintenance.

[0021] Furthermore, in such a design, the very short distance between the sample positioning system and the sample improves the optical conjugation between the sample positioning system and the focal plane within the sample, resulting in a scan angle that is closer to perpendicular, improving the signal.

[0022] Furthermore, this design allows for the construction of a conjugate device with a short footprint between the beam scanner and the sample for the best optical performance and signal. As a result, sample scanning can be performed very precisely, facilitating sample positioning. Additionally, a larger field of view (approximately 4-5 mm) is obtained.

[0023] Furthermore, because the sample position detection system is not coupled via the illumination arm of the full-field OCT imaging system, the signal degradation on the sample position detection system side described above does not occur.

[0024] According to other advantageous aspects of the invention, the imaging device comprises one or more of the following features, taken alone or in any technically possible combination:

[0025] The sample localization system is an optical sample localization system, and the detection axis (Δ) of the sample localization system is the optical axis.

[0026] The sample beam splitter is a dichroic mirror.

[0027] The optical system further comprises a reference beam splitter disposed in the reference arm between the primary beam splitter and the reference reflective surface, the reference beam splitter having the same optical properties as the sample beam splitter.

[0028] The reference beam splitter is identical to the sample beam splitter, and the reference beam splitter and the sample beam splitter have similar orientations relative to the propagation direction of the probe beam light waves propagating through them.

[0029] The optical sample localization system includes an OCT imaging system configured to determine, for each interferometric image acquired by the full-field OCT imaging system, at least one cross-sectional image of the sample at the time of acquisition of the interferometric image.

[0030] the OCT imaging system, - a secondary light source configured to output a position beam; a second-order interferometer, a secondary sample arm for receiving the sample, the optical axis of the secondary sample arm being the optical axis of the optical sample positioning system; a secondary reference arm consisting of a secondary reflector; a secondary beam splitter for separating the secondary sample arm and the secondary reference arm; Second-order interferometer with; a detector configured to output said position signal, wherein when said sample is placed in said secondary sample arm, said position signal is: A secondary reference wave obtained by reflection of the incident light wave of the position beam at the secondary reflector and A secondary sample wave obtained by scattering the initial portion of the location beam by a sample volume located along the propagation direction of the initial portion of the location beam. is representative of the resulting secondary optical interference signal resulting from interference between the detector: Includes.

[0031] the full-field OCT imaging system includes a first lens system disposed in the sample arm downstream of the primary beam splitter in a direction from the primary light source toward the primary beam splitter, and the OCT imaging system further comprises: a beam scanner including an optical deflection element for deflecting a position beam, the beam scanner being arranged in the secondary sample arm between the sample beam splitter and the secondary beam splitter; - a conjugation device comprising two lenses arranged in a 4F configuration, the back focal plane of the first lens system coinciding with a first focal plane of the conjugation device, and the optical deflection element being located in a second focal plane (F2) of the conjugation device; Includes:

[0032] Further comprising at least one compensation lens and / or deformable lens and / or variable focus lens disposed in the sample arm of the first-order interferometer.

[0033] The system further comprises a processing unit configured to determine at least one front image of the sample based on each acquired interference image and the corresponding position signal.

[0034] The processing unit is further configured to perform a denoising on each front image, the denoising comprising: filtering the front image using a Noise2Noise model; - acquiring at least one training en face image set using the full-field OCT imaging system, wherein each training en face image set includes a first en face image and a second en face image of the same section of the training sample at different acquisition times; For each training set, provide the Noise2Noise model with the first frontal image as input and the second frontal image as target output. This involves providing the noise to a pre-trained Noise2Noise neural network.

[0035] The present invention further provides a method for in vivo full-field coherent imaging of a scattering sample, the method comprising: - placing the sample in the sample arm of the full-field OCT imaging system as defined above; - acquiring at least one position signal representative of a position of said sample along a detection axis in time; - determining the position of the sample over time based on each position signal; - adjusting the position of the imaged slice of the sample based on the determined position of the sample.

[0036] According to other advantageous aspects of the invention, the imaging method comprises one or more of the following characteristics, taken alone or in any technically possible combination:

[0037] The sample is an in vivo eyeball of a subject.

[0038] The sample arm of the full-field OCT imaging system includes a first lens system configured to focus the probe beam onto the cornea or retina of the eye.

[0039] training the Noise2Noise neural network, - acquiring at least one training en face image set using the full-field OCT imaging system, wherein each training en face image set includes a first en face image and a second en face image of the same section of the training sample at different acquisition times; - for each training set, providing the first frontal image as an input and the second frontal image as a target output to a Noise2Noise model; moreover - determining at least one front image of the current sample based on each acquired interference image and the corresponding position signal; Each determined frontal image is provided to the trained Noise2Noise neural network for denoising. [Brief explanation of the drawings]

[0040] [Figure 1] 1 is a schematic diagram of a first embodiment of an imaging device according to the present invention. [Figure 2] FIG. 2 is a schematic diagram of a 4F conjugate device of the imaging apparatus of FIG. 1. [Figure 3] 3A and 3B are en face images of the corneal stroma and spongiosum, respectively, acquired by the imaging device of FIG. 1, and FIG. 3C is a cross-sectional image of the cornea acquired by the imaging device of FIG. [Figure 4] 2A and 2B are examples of front images before and after a noise removal step executed by a processing unit of the imaging device of FIG. 1. [Figure 5] FIG. 2 is a schematic diagram of a second embodiment of an imaging device according to the present invention. [Figure 6] FIG. 10 is a schematic diagram of a third embodiment of an imaging device according to the present invention. [Figure 7] 7A and 7B are schematic diagrams of a fourth embodiment of an imaging device according to the present invention. [Figure 8] 10 is a schematic diagram of a fifth embodiment of an imaging device according to the present invention.The present invention will be better understood with reference to the accompanying drawings, in which: Detailed Description of the Invention

[0041] First embodiment An imaging device 2 according to the present invention is shown in FIG.

[0042] The imaging device 2 is particularly suitable for in vivo imaging of a scattering sample 4, in particular a moving sample, typically a subject's eye. More precisely, the imaging device 2 is configured for full-field interferometric imaging of the sample 4.

[0043] In particular, the imaging device 2 is configured to determine at least one en-face image (and preferably at least one cross-sectional image) of at least a portion of a sample 4, such as the cornea 6 of an in-vivo eye 4. Preferably, the imaging device 2 is also configured to provide at least one 3D image of at least a portion of the sample 4, e.g. based on the determined en-face and cross-sectional images.

[0044] The imaging device 2 is composed of a full-field OCT imaging device 8 (also called a “FF-OCT imaging device”), an optical sample position detection device 10, and a processing device 12.

[0045] The FF-OCT imaging system 8 is configured to provide at least one interferogram representative of a front view of the sample 4 , said front view being calculated by a processing unit 12 .

[0046] According to the present invention, the front image of the sample 4 is an image of a depth cross section of the sample 4. Such a depth cross section corresponds to a cross section of the sample 4 taken along a plane perpendicular to the optical axis ZZ of the FF-OCT imaging system 8.

[0047] Furthermore, the optical sample localization system 10 is configured to output a position signal representative of the position of the sample 4 along the optical axis ZZ. The position signal is, for example, an optical spectrum or a time signal having variations representative of the movement of the sample.

[0048] Preferably, the processing unit 12 is also configured to calculate at least one cross-sectional image of at least a portion of the sample 4 based on the position signal.

[0049] Preferably, the processing unit 12 is also configured to calculate at least one 3D image of at least a part of the sample 4 based on the time sequence of the en face images and the associated cross-sectional images.

[0050] Preferably, the imaging device 2 further comprises a main movable platform 13 (e.g., a motorized platform or a manual platform manually controlled by an operator) for mounting the FF-OCT imaging system 8 and the optical sample positioning system 10, the main movable platform being movable along three independent (e.g., perpendicular) directions so as to jointly translate the FF-OCT imaging system 8 and the optical sample positioning system 10.

[0051] FF-OCT Imaging System 8 The FF-OCT imaging system 8 will now be described.

[0052] The FF-OCT imaging system 8 consists of a primary light source 14 , a primary interferometer 16 , a camera 18 , and a sample beam splitter 19 .

[0053] Primary light source 14 The primary light source 14 is configured to output a probe beam 20 (shown as a dashed line in the figure) and is positioned at the input of the primary interferometer 16 .

[0054] Preferably, the spectrum of the probe beam 20 has a central peak at wavelengths belonging to the interval [0.4 μm;1 μm], more preferably the interval [0.75 μm;1 μm]. Sudachi

[0055] The primary light source 14 is a light source with low spatial coherence (on the order of half the wavelength, e.g., less than 1.5 μm when using a second lens system 36 with a numerical aperture of 0.3) and low temporal coherence length (e.g., less than 20 μm). For example, the primary light source 14 can be a thermal light source (such as a halogen lamp), a light-emitting diode, or a mode-mixing multimode fiber followed by a laser.

[0056] Preferably, an optical element 22 is placed at the output of the primary light source 14 to provide Kohler illumination.

[0057] The primary light source 14 is positioned so that the probe beam 20 emitted by the primary light source 14 reaches a primary beam splitter 24 (described below) of the primary interferometer 16, whereby a first portion of the probe beam 20 propagates in a sample arm 26 of the primary interferometer 16 and a second portion of the probe beam 20 (different from the first portion of the probe beam 20) propagates in a reference arm 28 of the primary interferometer 16.

[0058] Primary Interferometer 16 The primary interferometer 16 includes the aforementioned primary beam splitter 24, a sample arm 26, and a reference arm 28. The sample arm 26 is for receiving the sample 4.

[0059] The first-order interferometer 16 is configured to generate a first-order optical interference signal resulting from interference occurring at each point in the imaging field of the FF-OCT imaging system 8 when the sample 4 is placed in the sample arm 26 between a sample wave obtained from the sample arm 26 and a reference wave received from the reference arm 28. The sample wave and the reference wave are described further below.

[0060] Due to the low temporal coherence of the primary light source 14, interference only occurs when the sample arm 26 and reference arm 28 of the primary interferometer 16 are aligned with micron accuracy (typically 1-10 microns). In this embodiment, when the sample 4 is placed in the sample arm 26 of the primary interferometer 16, interference only occurs for the reflected light from a sample layer that is a few microns thick. The reflected light from other sample layers does not interfere with the reference light coming from the reference arm 28.

[0061] Preferably, the first order interferometer 16 is a Linnik interferometer, as shown in Figure 1. However, the first order interferometer 16 may be any other suitable type of interferometer, such as a Michelson interferometer.

[0062] Each element of the first-order interferometer 16 will now be described in more detail.

[0063] Primary Beam Splitter 24 The primary beam splitter 24 separating the sample arm 26 and the reference arm 28 is, for example, a non-polarizing splitter cube.

[0064] Sample arm 26 As mentioned above, the sample arm 26 is for receiving the sample 4. The sample arms have respective optical axes that define the optical axis ZZ of the FF-OCT imaging system 8.

[0065] The sample arm 26 is positioned so that the sample 4 receives the aforementioned first portion of the probe beam 20 from the primary beam splitter 24 .

[0066] Furthermore, the aforementioned sample wave is obtained by scattering of the incident light wave of the first portion of the probe beam 20 by the sample 4. More precisely, at each point of the imaging field, a corresponding sample wave is obtained by scattering (e.g., backscattering) of the incident light wave of the first portion of the probe beam 20 by an elementary volume of a slice of the sample 4, the elementary volume being conjugate to the point of the imaging field, and all elementary volumes having the same predetermined depth, i.e., the same coordinate along the optical axis ZZ.

[0067] Reference arm 28 The reference arm 28 has an optical axis XX and defines a reference reflecting surface 30 .

[0068] The reference arm 28 is positioned such that the reference reflective surface 30 receives the aforementioned second portion of the probe beam 20 from the primary beam splitter 24, distinct from the first portion.

[0069] In this case, said reference wave is obtained by reflection of the incident light wave of said second part of the probe beam 20 on the reference reflecting surface 30. More precisely, at each point of the imaging field, a corresponding reference wave is obtained by reflection of the incident light wave of said second part of the probe beam 20 on a fundamental plane of the reference reflecting surface 30 that is conjugate to said point of the imaging field.

[0070] Preferably, the reference reflective surface 30 is flat. For example, the reference reflective surface 30 is a metal mirror, a neutral density filter glass, or a glass plate.

[0071] Alternatively, the reference reflective surface 30 is curved, preferably with a curvature that matches the convex curvature of the cornea. This feature is advantageous because it provides an en face image that represents all the cellular layers of the cornea at once, rather than lying in a plane perpendicular to the ZZ axis.

[0072] Advantageously, the reference reflecting surface 30 is mounted on a piezoelectric stage 32. Such a piezoelectric stage 32 can be used for phase modulation, as will be explained later in relation to the processing unit 12. Preferably, the piezoelectric stage 32 is configured to operate in the range [100 Hz; 10 kHz].

[0073] Camera 18 A camera 18 is positioned at the output of the first-order interferometer 16 .

[0074] The camera 18 is configured to acquire at least one interference image representative of the aforementioned primary optical interference signal resulting from interference in the imaging field between the reference wave and the sample wave when the sample 4 is placed in the sample arm 26.

[0075] Each interference image is a two-dimensional image, and once processed by the processing unit 12, a front image of the sample 4 can be obtained, i.e., an image of the sample in a plane perpendicular to the optical axis ZZ of the sample arm 28 and conjugate to the imaging field.

[0076] Preferably, the camera 18 is configured to acquire interference images at high speed, for example at a frequency between 100 Hz and 1 kHz or higher, depending on the dynamics of the sample 4, and more particularly the dynamics of the movements within the sample 4.

[0077] For example, the camera 18 is a charge-coupled device (CCD) camera or a complementary metal-oxide semiconductor (CMOS) camera.

[0078] Sample beam splitter 19

[0079] The sample beam splitter 19 is arranged in the sample arm 26 and intersects the optical axis ZZ of the sample arm 26. More precisely, the sample beam splitter 19 is arranged downstream of the primary beam splitter 24 in the direction from the primary light source 14 to the primary beam splitter 24. That is, as shown in FIG. 1, when the sample 4 is arranged in the sample arm 26, the sample beam splitter 19 is arranged between the primary beam splitter 24 and the sample 4.

[0080] The sample beam splitter 19 is preferably configured to transmit light having wavelengths within a predetermined range centered around the central wavelength of the primary light source 14 and to reflect light having wavelengths within a predetermined range centered around the central wavelength of a secondary light source 46 (described below) of the optical sample localization system 10. For example, the sample beam splitter 19 is a dichroic mirror.

[0081] Additional optional optical elements Reference Beamsplitter 33 Advantageously, the primary interferometer 16 also includes a reference beam splitter 33 located in the reference arm 28 between the primary beam splitter 24 and the reference reflective surface 30. This is advantageous because such a reference beam splitter 33 introduces a dispersion in the reference arm 28 similar to the dispersion caused by the sample beam splitter 19 in the sample arm 26, thereby improving the ability to extract a front image from the interference image acquired by the FF-OCT imaging system 8.

[0082] Reference beam splitter 33 has the same optical properties as sample beam splitter 19. For example, reference beam splitter 33 is identical to sample beam splitter 19.

[0083] Preferably, reference beam splitter 33 and sample beam splitter 19 are oriented so that they subtend the same angle with the direction of propagation of the light waves of probe beam 20 propagating through them. In other words, reference beam splitter 33 and sample beam splitter 19 are oriented so that the angle between the plane in which sample beam splitter 19 extends and the direction of propagation of the first portion of probe beam 20 is equal to the angle between the plane in which reference beam splitter 33 extends and the direction of propagation of the second portion of probe beam 20.

[0084] More precisely, for each point in the field of the probe beam 20: - on the one hand, the angle between the propagation direction of the first portion of the probe beam 20 and the plane in which the principal surface of the sample beam splitter 19 lies, onto which the light emitted from said point of the illumination field is incident; on the other hand, the angle between the propagation direction of the second portion of the probe beam 20 and the base plane of the reference beam splitter 33 on which the light emitted from said point of the illumination field is incident; are essentially equal.

[0085] This feature is advantageous because the FF-OCT imaging system 8 can be sensitive to small mismatches between the sample arm 26 and the reference arm 28 (contrary to conventional SD-OCT systems). Indeed, by matching the angles between the identical beam splitters 19, 33 of the sample arm 26 and the reference arm 28, the dispersion of the arms 26, 28 of the primary interferometer 16 is balanced. As a result, with proper adjustment, the optical path length difference between the sample arm 26 and the reference arm 28 becomes smaller than the coherence length (e.g., 8 μm) of the primary light source 14, thereby enabling the FF-OCT imaging system 8 to acquire reliable interference images.

[0086] This feature also allows for lateral (i.e., perpendicular to the propagation of light) optical alignment in the sample arm 26 and the reference arm 28, thereby further improving the ability of the FF-OCT imaging system 8 to acquire reliable interference images.

[0087] First and second lens systems 34, 36 Preferably, the first-order interferometer 16 further comprises a first lens system 34 and a second lens system 36 disposed in each of the sample arm 26 and reference arm 28, respectively.

[0088] More precisely, the first lens system 34 is arranged such that, when the sample 4 is placed in the sample arm 26, the focal plane of the first lens system 34 coincides with the cross-section of the sample 4 to be imaged. Furthermore, the first lens system 34 is arranged such that the corresponding optical axis coincides with the optical axis ZZ of the sample arm 26.

[0089] In practice, the depth of focus of the first lens system 34 is always larger along the optical axis ZZ than the slice width that actually causes interference due to the low temporal coherence of the first light source 14. In other words, the camera 18 captures light from a relatively thick region of the sample 4, but the optical interference signal that appears in the front image is located in a thin region of the depth of focus, the thickness of which is determined by the primary light source 14. By acquiring multiple camera frames with different phases, it is possible to extract only the thin interference fringe portion.

[0090] Preferably, first lens system 34 is located downstream of sample beam splitter 19 in the direction from primary beam splitter 24 towards sample beam splitter 19 .

[0091] Furthermore, the second lens system 36 is positioned so that the reference reflective surface 30 is located in the focal plane of the second lens system 36 .

[0092] Furthermore, the second lens system 36 is positioned so that its corresponding optical axis coincides with the optical axis XX of the reference arm 28 .

[0093] Advantageously, each lens system 34, 36 has a high numerical aperture (typically about 0.3) to ensure cellular resolution while providing a large field of view (typically about 1 mm. For example, each of the first lens system 34 and the second lens system 36 is a microscope objective.

[0094] Advantageously, the reference reflective surface 30 and the second lens system 36 are mounted on a motorized platform 40 movable along the optical axis XX of the reference arm 28 .

[0095] This is advantageous because the displacement of the motorized platform 40 along the optical axis XX allows the imaging slice to be positioned based on the position of the sample. More precisely, when imaging the retina of the eye 4, the displacement of the motorized platform 40 along the optical axis XX allows the depth of the imaging slice (i.e., the plane orthogonal to the optical axis ZZ that is imaged by the FF-OCT imaging system 8) to be selected. Furthermore, when imaging the cornea, the displacement of the motorized platform 40 allows the correction of the defocus in the eye caused by Snell's law and dependent on the position of the eye relative to the FF-OCT imaging system 8, so that the imaged slice falls within the current focal plane.

[0096] Primary Spectral Filter 42 Preferably, the first-order interferometer 16 further comprises a first-order spectral filter 42 positioned between the output of the first-order interferometer 16 and the camera 18 .

[0097] The primary spectral filter 42 is configured to transmit light having wavelengths within a predetermined range centered around the central wavelength of the primary light source 14 and to block light having wavelengths outside the predetermined range, such as light emitted from the optical sample localization system 10. This prevents the camera 18 from detecting photons emitted from the localization system 10.

[0098] For example, the first-order spectral filter 42 is an optical bandpass filter.

[0099] Third Lens System 44 Preferably, the first-order interferometer 16 also includes a third lens system 44 , such as an achromatic doublet, positioned between the output of the first-order interferometer 16 and the camera 18 .

[0100] The third lens system 44 has a focal length selected to make the detection plane of the camera 18 conjugate with the plane located at the focal point of the first and second optical systems 34, 36. For example, the focal length of the third lens system 44 is several hundred millimeters, typically 300 millimeters.

[0101] Such a third lens system 44 (typically a tube lens) can also be used, and advantageously, to change the magnification and field of view of the camera 18. In practice, the first and second lens systems 34, 36 are typically "infinity corrected." For example, a Nikon microscope objective 34 (with a tube lens focal length equal to 200 mm) paired with a 200 mm tube lens 44 provides a 10x magnification field of view, while a 300 mm tube lens 44 provides a 15x magnification field of view.

[0102] Tunable Lens 45 Preferably, the first-order interferometer 16 further includes an adjustable (ie, adaptive) lens 45 disposed in the sample arm 26 .

[0103] The tunable lens 45 is operable by the processing unit 12, advantageously in combination with the motorized platform 40, to correct for defocusing on the part of the FF-OCT imaging system due to the fact that the eye does not have the same refractive index as air. More precisely, the tunable lens 45 is operable to control the focal position of the initial part of the probe beam 20 that impinges on the sample 4.

[0104] Optionally, adjustable lens 45 may be configured to provide a similar function as third lens system 44, in which case third lens system 44 may be removed.

[0105] Optical sample position measurement system 10 Next, the location system 10 will be described.

[0106] As previously mentioned, the optical sample positioning system 10 (also referred to as the "positioning system") is configured to output a position signal representative of the position of the sample 4 along the optical axis ZZ.

[0107] The localization systems 10 are disposed outside the sample arm 26 and have respective optical axes Δ that intersect with the sample beam splitter 19. Due to the presence of the sample beam splitter 19, the positioning beam 51 emitted by the positioning system 10 along its optical axis Δ is reflected by the sample beam splitter 19 to propagate substantially parallel to the optical axis ZZ of the FF-OCT imaging system 8. Thus, after reflection by the sample beam splitter 19, the optical axes Δ and ZZ are coaxially aligned.

[0108] It is noted that the propagation direction of the positioning beam 51 (i.e., the optical axis Δ) is truly parallel to the optical axis ZZ only for a single orientation of the deflection element 74 (described below) of the positioning system 10. For all other positions of the deflection element 74 (all of which are used during acquisition of position signals leading to two-dimensional cross-sectional images of the sample 4), there is an angle between the optical axis Δ and the optical axis ZZ.

[0109] Conversely, light (of the appropriate wavelength) from the sample 4 can reach the position detection system 10 after being reflected by the sample beam splitter 19 .

[0110] Preferably, the localization system 10 includes an OCT imaging system, as described below, such as a spectral domain OCT system (also called an "SD-OCT system"), a swept-source OCT system, or a time-domain OCT system.

[0111] In this case, the localization system 10 includes the secondary light source 46, the second-order interferometer 48, and the detector 50 previously described.

[0112] Preferably, the location system 10 further includes a conjugation device 53 shown in FIG.

[0113] Secondary light source 46 The secondary light source 46 is configured to output the aforementioned position beam 51 (dotted line in the drawing), and is disposed on the input side of the second-order interferometer 48 .

[0114] Advantageously, the position beam has a spectrum that does not at least partially overlap with the spectrum of the light emitted by the primary light source 14 .

[0115] For example, the spectrum of the position beam 51 has a central peak at a wavelength belonging to the [870 nm; 1000 nm] interval, while the primary light source 14 emits light in the [750 nm; 860 nm] interval.

[0116] Preferably, the secondary light source 46 is a spatially coherent light source, for example, a superluminescent diode or a swept laser source (when using a swept-source OCT system).

[0117] Preferably, when the positioning system 10 includes an SD-OCT system, the secondary light source 46 has low temporal coherence (i.e., a broad spectrum compared to its central wavelength). Furthermore, in the case of a swept laser light source, at each instant the positioning beam has a fairly large temporal coherence, but is swept over a broad spectrum.

[0118] The secondary light source 46 is coupled to the input of a second-order interferometer 48 via, for example, a first optical fiber 52 .

[0119] The secondary light source 46 is positioned so that the position beam 51 emitted by the secondary light source 46 reaches a secondary beam splitter 54 (described below) of the secondary interferometer 48, whereby a first portion thereof propagates in a secondary sample arm 56 of the secondary interferometer 48 and a second portion thereof (different from the first portion of the position beam 51) propagates in a secondary reference arm 58 of the secondary interferometer 48.

[0120] Second-order interferometer 48 The secondary interferometer 48 includes the secondary beam splitter 54, secondary sample arm 56, and secondary reference arm 58 previously described.

[0121] The secondary beam splitter 54 separates a secondary sample arm 56 and a secondary reference arm 58. Further, the secondary sample arm 56 is for receiving the sample 4.

[0122] The secondary interferometer 48 is configured to generate a secondary optical interference signal resulting from interference between a secondary sample wave obtained from the secondary sample arm 56 and a secondary reference wave received from the secondary reference arm 58 when the sample 4 is placed in the secondary sample arm 56. The secondary sample wave and the secondary reference wave will be described later.

[0123] Due to the structure of the imaging device 2, placing the sample 4 in the sample arm 26 of the primary interferometer 16 is equivalent to placing the sample 4 in the secondary sample arm 56 of the secondary interferometer 48.

[0124] Secondary Reference Arm 58 Preferably, the secondary reference arm 58 includes an optical fiber 60, an optional lens 62, an optional dispersion compensation plate 64, and a secondary reflector 66 (such as a metallized mirror) arranged in series.

[0125] In this case, the incident light wave of the second portion of the location beam is reflected by the secondary reflecting mirror 66 to obtain the aforementioned secondary reference wave.

[0126] Advantageously, the length of the optical fiber 60 is selected so that the optical path between the secondary beam splitter 54 and the secondary reflecting mirror 66 is equal to the optical path between the secondary beam splitter 54 and the cell layer in the sharp focus of the first lens system 34, i.e., the cell layer currently being imaged. Preferably, the length of the optical fiber 60 is selected so that the optical path between the secondary beam splitter 54 and the secondary reflecting mirror 66 is equal to the optical path between the secondary beam splitter 54 and the predetermined position where the sample 4 is expected to be located during operation of the imaging device 2.

[0127] Secondary sample arm 56 Preferably, the secondary sample arm 56 includes an optical fiber 68, a beam scanner 70 and a secondary spectral filter 72 arranged in series.

[0128] In this case, the aforementioned secondary sample wave is obtained by scattering the incident light wave of the first portion of the location beam by the sample 4. More precisely, the sample wave is obtained by scattering the incident light wave of the first portion of the location beam by an elementary volume of the sample 4 located along the propagation direction of the first portion of the location beam after deflection by the beam scanner 70.

[0129] The beam scanner 70 is arranged between the sample beam splitter 19 and the secondary beam splitter 54. The beam scanner 70 is configured to scan the location beam in a plane perpendicular to the optical axis XX. More precisely, the beam scanner 70 includes a deflecting element 74 (typically a mirror) for deflecting the location beam 51. As a result, upon reflection of the location beam at the sample beam splitter 19, operation of the beam scanner 70 (i.e., movement of the deflecting element 74) results in scanning of the location beam at the sample 4 in a plane perpendicular to the optical axis ZZ, more precisely, scanning of the location beam along a plurality of lines (each line corresponding to the propagation direction of the initial part of the location beam 51) in the plane perpendicular to the optical axis ZZ.

[0130] For example, the beam scanner 70 is configured to scan one line approximately every 10 μs. In other words, the beam scanner 70 is configured to scan the eye 4 at a time interval shorter than the time constant of the eye position fluctuation, i.e., a time shorter than the typical time during which the eye position can fluctuate over a large range (10 μm or more), typically at a time interval of about 10 ms.

[0131] The secondary spectral filter 72 is configured to transmit light having wavelengths within a predetermined range centered on the central wavelength of the secondary light source 46 and to block light having wavelengths outside the predetermined range, for example, light emitted from the FF-OCT imaging device 8. For example, the secondary spectral filter 72 is an optical bandpass filter. This prevents the detector 50 from detecting photons emitted from the FF-OCT imaging system 8.

[0132] Detector 50 A detector 50 is coupled to the output of the second-order interferometer 48 .

[0133] Detector 50 is configured to detect the secondary optical interference signal and output a corresponding position signal over time.

[0134] For example, if the localization system 10 is a time-domain OCT system such as a swept-source OCT system, the detector 50 is a photodetector such as a photodiode, or if the localization system 10 is a spectral-domain OCT system, the detector 50 is a spectrometer.

[0135] Conjugation Device 53 Conjugation device 53 includes two lenses in a 4F configuration as shown in FIG.

[0136] More precisely, the conjugator 53 is positioned so that the rear focal plane of the first lens system 34 coincides with the first focal plane F1 of the conjugator 53, and the deflection element 74 of the beam scanner 70 is positioned in the second focal plane F2 of the conjugator 53.

[0137] Due to the presence of the conjugate element 56, the positioning beam 51 is scanned by the beam scanner 70 in a plane perpendicular to the optical axis XX and is located at approximately the same depth as the imaging field determined by the first lens system 34. This is illustrated by the cross and dotted beams in FIG. 3, which correspond to the paths of the positioning beam 51 at two different deflection angles of the deflection element 74. As a result, the deflection element 74 is located substantially in the back focal plane of the first lens system 34. As a result, even if the beam scanner 70 is located away from the back focal plane, the problem of only light propagating at a small angle relative to the propagation direction of the probe beam entering the entrance pupil of the first lens system 34 is solved. Furthermore, lenses with a short focal length (approximately 3 cm) can be used, resulting in a very compact conjugate device 53.

[0138] Differences between the FF-OCT imaging system 8 and the SD-OCT version of the optical specimen position detection system 10 As mentioned above, due to the low temporal coherence of the primary light source 14, interference phenomena only occur when the sample arm 26 and reference arm 28 of the primary interferometer 16 are matched with micron accuracy (typically 1-10 microns).

[0139] At first glance, due to the low temporal coherence of the secondary light source 46 in an SD-OCT system, it might seem that interference phenomena occur only when the secondary sample arm 56 and secondary reference arm 58 of the secondary interferometer 48 are aligned with micron accuracy (as in an FF-OCT system). However, the inventors discovered that this is not the case in the SD-OCT version of the optical sample localization system 10, in which the detector is a spectrometer consisting of a diffraction grating and a detector. In this case, the diffraction grating separates the broadband, low-temporal-coherence spectrum of the reflected secondary light source 46 (coming from the secondary sample arm 56 and secondary reference arm 58, respectively) into thin, highly temporally coherent bands. Each of these bands is then detected by a separate detector. Therefore, even if the secondary sample arm 56 and secondary reference arm 58 of the interferometer are significantly misaligned (on the order of 1 to 10 millimeters), interference appears to occur at each detector.

[0140] To reconstruct cross-sectional images, SD-OCT relies on the fact that interference frequencies vary for different mismatch distances. The interference frequency is small when the secondary sample arm 56 and the secondary reference arm 58 are closely matched, but increases as the mismatch increases. By Fourier transforming the measured spectrum, all frequencies can be extracted and converted to distances. As a result, SD-OCT can reconstruct detailed one-dimensional profiles (A-lines) of scatterers inside the sample.

[0141] A two-dimensional cross-sectional image can be reconstructed by placing a beam scanner in the secondary sample arm 56, such as the deflection element 74. Note that there is no scanning in the secondary reference arm 58, and the different secondary sample beams (coming from the secondary sample arm 56) interfere with the single secondary reference beam (coming from the secondary reference arm 58).

[0142] Therefore, SD-OCT is a point-by-point imaging system that is not affected by wavefront misalignment of the secondary sample arm 56 and secondary reference arm 58, in contrast to FF OCT systems 8 that simultaneously capture all pixels of a frontal tomographic image.

[0143] Furthermore, we have found that the SD-OCT and FF-OCT systems 8 are not similarly affected by the insertion of dichroic mirrors, such as the sample beam splitter 19, into the sample arm 26 and secondary sample arm 56. For example, in the case of a dichroic mirror consisting of a 1 mm glass plate, the glass plate introduces an optical path difference of 1 * 1.5 - 1 * 1 = 0.5 mm (thickness * refractive index) between the sample (respectively secondary sample) arm and the reference (respectively secondary reference) arm. The SD-OCT system still exhibits interference, and the final image appears unaffected except for some dispersion, which leads to depth shifts and reduced axial resolution. Thus, the wavefront of the SD-OCT system does not require correction because it is a point-image system.

[0144] On the other hand, the FF-OCT system 8 requires alignment of the optical arm with a precision of a few microns, and the inventors found that a mismatch of 0.5 mm appeared to destroy all interferometric signals. It was not possible to re-image. Furthermore, because the FF-OCT system 8 is a full-field, not point-by-point imaging system, it suffers from wavefront mismatch.

[0145] Furthermore, the inventors have discovered that glass plates placed in the FF-OCT system 8 introduce wavefront curvature. One way to compensate for optical misalignment, dispersion, and wavefront misalignment in FF-OCT is to insert an identical dichroic mirror (described later in a second embodiment) in the reference arm 28 as in the sample arm 26 and rotate it to the exact same angle.

[0146] Because it is not easy to obtain the exact same angle with micron-level accuracy, an additional alignment step is required. First, the FF-OCT system 8 interferometer is aligned in the usual way, but with two dichroic mirrors 19 and 33, positioned in the sample arm 26 and reference arm 28, perpendicular to the light path. After obtaining an interference image on the screen, one of the dichroic mirrors 19 or 33 is tilted. As a result, the interference signal disappears due to the mismatch. Next, the second dichroic mirror 33 or 19, respectively, is tilted to recover the signal. This process continues step by step: one dichroic mirror is tilted slightly, followed by another dichroic mirror, to recover the signal. Final alignment is achieved when both dichroic mirrors 19 and 33 are tilted at 45 degrees.

[0147] It should be noted that with respect to the FF-OCT system of WO2020 / 065068A1, the signal loss issue does not apply since the dichroic mirror is located in the illumination arm (i.e., probe beam 20), and therefore the sample arm and reference arm remain aligned. Due to the increased sensitivity of FF-OCT systems compared to SD-OCT systems, alternative configurations that solve the technical problems cited in the background of the present invention would not have been possible prior to the present invention.

[0148] Processing Unit 12 According to the present invention, the term "processing unit" should not be construed as being limited to hardware capable of executing software, but generally refers to a processing device that may include, for example, a microprocessor, an integrated circuit, or a programmable logic device (PLD). A processing device may also encompass one or more graphics processing units (GPUs) or tensor processing units (TSUs), whether utilized for computer graphics and image processing or other functions. Furthermore, instructions and / or data enabling the execution of the associated and / or resulting functions may be stored on any processor-readable medium, such as, for example, an integrated circuit, a hard disk, an optical disk such as a compact disc (CD) or a digital versatile disc (DVD), a random-access memory (RAM), or a read-only memory (ROM). Instructions may be stored in hardware, software, firmware, or any combination thereof.

[0149] The processing unit 12 is configured to calculate each in-plane image based on each interference image output by the FF-OCT imaging device 8.

[0150] To this end, the processing unit is configured to control the piezoelectric stage 32 to displace the reference reflecting surface 30 along the optical axis XX, thereby modulating the reference wave, with the amplitude of such modulation being set according to a predetermined phase demodulation scheme, for example, a two-phase demodulation scheme or a four-phase demodulation scheme.

[0151] For example, in the two-phase scheme, the processing unit 12 is configured to determine the front image from two interference images acquired by the camera 18: a first interference image is captured with the reference reflective surface 30 in a first position, and a second interference image is captured with the reference reflective surface 30 shifted by an optical phase of π, i.e., half the central wavelength of the probe beam 20.

[0152] As mentioned earlier, interference only occurs for light reflected from sample layers a few microns thick. Light reflected from other sample layers does not interfere with the reference light from the reference arm 28. By temporally modulating the reference reflecting surface 30 in this way, multiple interference patterns (optical phases) can be acquired. In the simplest demodulation scheme, two opposite (π phase shifted) interference patterns are subtracted, eliminating the incoherent light from other sample layers and doubling the signal from the layer of interest (tomography).

[0153] As another example, in a four-phase scheme, processing unit 12 is configured to determine a front image from four interference images acquired by camera 18, where the images are captured sequentially. Assuming the phase of the first frame is 0, processing unit 12 is configured to control piezoelectric stage 32 and camera 18 such that the phase shifts of subsequent frames relative to the first frame are π / 2, π, and 3π / 2.

[0154] The processing unit 12 is further configured to calculate the reflectance of sample structures within the coherence volume (i.e., plane of interest) based on the pairwise subtraction of the phase-shifted images. The resulting en face image (an example of which is shown in FIG. 3A) represents the calculated reflectance values ​​across the plane of interest. More precisely, FIG. 3A shows healthy corneal stroma of a human eye.

[0155] Additionally, Figure 3B shows the trabecular meshwork, a region at the ocular periphery located where the corneal endothelium and iris meet. Figure 3B demonstrates that the imaging device 2 is capable of providing images with subcellular resolution of this small, difficult-to-locate region (approximately 0.5 mm in length). This is achieved thanks to the imaging device's ability to accurately track the position of the eye 4 through optimized use of the positioning system 10.

[0156] Glaucoma is a vision-destroying disease that is extremely difficult to diagnose early. The ability to image the cavernous network at high resolution (unavailable with known imaging systems) could allow for early detection of glaucoma and prevent the risk of vision loss.

[0157] Alternatively, or in addition, for a sample that moves along the ZZ axis (typically an eye that exhibits natural physiological eye movements), the processing unit 12 is configured to calculate each circular face image based on each interference image output by the FF-OCT imaging system 8 without modulating the position of the reference reflective surface 30. In this case, slight movements of the sample along the ZZ axis result in interference images that are out of phase with respect to each other, thereby allowing the calculation of a frontal image.

[0158] Preferably, the processing unit 12 is also configured to control the position of the motorized platform 40 based on the position signal to compensate for sample movement so that the same sample slice is imaged over time.

[0159] More precisely, for a given position of the deflection element 74, the detector 50 acquires a position signal representative of a line corresponding to the propagation direction of the first portion of the position beam 51. The processing unit 12 is typically configured to reconstruct the portion of the sample intersected by said line.

[0160] Preferably, the processing unit 12 further controls the beam scanner 70 to acquire position signals for different positions (i.e., different adjacent lines) of the deflection element 74. The processing unit 12 is further configured to calculate at least one cross-sectional image of at least a portion of the sample 4 based on the plurality of position signals. An example of such a cross-sectional image of the cornea is shown in Figure 3C.

[0161] Furthermore, the processing unit 12 is configured to segment the cross-sectional images so as to identify the position of at least one predetermined structure (such as a cellular structure) of the sample 4. As a result, by comparing each segmented cross-sectional image {right arrow over (x)} with a predetermined target center position {right arrow over (x)}, the processing unit 12 can determine the position of each segmented structure over time along the optical axis ZZ.

[0162] Preferably, the processing unit 12 is further configured to perform image processing on each computed en face image in order to correct for noise resulting from the operation of the FF-OCT imaging system 8 .

[0163] In fact, face images are generally noisy, as shown in Figure 4A, which mainly includes Gaussian noise, interference fringe noise, and background noise.

[0164] Gaussian noise originates from the detector, i.e., the camera 18. This type of noise is not specific to FF-OCT and is present in any image taken by the camera. Gaussian noise is uncorrelated, meaning that the noise in one pixel is independent of the noise in other pixels.

[0165] Interference fringe noise, visible in Figure 4A as alternating fringes of low and high intensity, results from the fact that due to movement of the sample 4 (like an eye), random phase fluctuations are introduced into the interference image acquired by the camera 18, in addition to those caused by the piezoelectric stage 132. These fluctuations result in the position and frequency of the interference fringes being random over time. Furthermore, because these phase fluctuations are random, they cannot be simply removed using, for example, a four-phase demodulation scheme, and the interference fringes will remain.

[0166] Interference fringes are correlated: if one pixel is very bright, nearby pixels are likely to be bright as well.

[0167] To perform noise correction, the processing unit 12 is configured to implement an artificial intelligence model, such as the Noise2Noise deep learning model.

[0168] More precisely, the model is pre-trained using a training image set, e.g., in vivo human corneal FF-OCT images (stroma and endothelium, etc.).

[0169] More precisely, at least one training set of en face images of the training sample 4 is acquired using the full-field OCT imaging system 10. Each training set includes a first en face image and a second en face image of the same section of the training sample 4, acquired at different acquisition times. Due to the high speed of the camera 18 (approximately 500 frames per second), the en face images can be viewed as if the eye movement were frozen (i.e., in the same position) for several frames. In this case, in each training set, the first and second en face images exhibit different noises. This is due to, on the one hand, Gaussian noise (which varies from one camera frame to the next) and, on the other hand, interferometric noise (which results in phase variations from one acquired interferometric image to another due to imperfect demodulation schemes and / or submicron eye movements between frames).

[0170] Then, to train the Noise2Noise model, for each training set, the first face image is provided as input and the second face image is provided as target output.

[0171] This is an advantage. Indeed, during such training, the Noise2Noise model tries to learn how to generate one specific noise pattern from another. Because the noise is random, the model cannot efficiently learn this transformation and ends up learning to transform the noisy pattern into the average of all noisy patterns, i.e., a clean image. Thanks to this training, both spatially uncorrelated noise (Gaussian) and correlated noise (interference fringes) are removed, as shown in Figure 4B, the result of denoising the face image in Figure 4A. This is due to the model's ability to learn the statistics of the correlated noise as well as the statistics of typical structures of the specimen visible in the image (e.g., eye structures).

[0172] operation Next, the operation of the imaging device 2 will be described.

[0173] In the setup step, the FF-OCT imaging system 8 and the optical sample positioning system 10 are aligned.

[0174] Furthermore, during a configuration step, the modulation / demodulation scheme is preferably selected by an operator and stored in the processing unit 12 .

[0175] Preferably, if the processing unit 12 is intended to perform noise correction, the configuration of the Noise2Noise neural network is performed during a configuration step, more precisely, the neural network is trained, as described above.

[0176] The sample 4 is then placed on the sample arm 26 of the FF-OCT imaging system 8 and the secondary sample arm 56 of the optical sample positioning system 10 .

[0177] The processing unit 12 then controls the piezoelectric stage 32 to displace the reference reflecting surface 30 along the optical axis XX, thereby modulating the reference wave. The processing unit 12 also controls the camera 18 to acquire an interference image according to a predetermined modulation scheme, i.e., to acquire an interference image when the reference reflecting surface 30 is at a position that causes a predetermined phase shift in the reference wave.

[0178] The processing unit 12 also controls the beam scanner 70 so that the location beam 51 scans the sample 4. The processing unit 12 simultaneously controls the detector 50 so that a location signal representative of each line currently scanned by the location beam 51 is output.

[0179] Then, the processing unit 12 calculates each front image based on each interference image output by the FF-OCT imaging device 8.

[0180] The processing unit 12 also calculates a cross-sectional image of the sample 4 based on the plurality of position signals and segments the calculated cross-sectional image to identify the location of at least one predetermined structure of the sample 4.

[0181] The processing unit 12 further compares the position of the divided cross-sectional image with a predetermined target center position to determine whether the position of each divided structure changes over time.

[0182] If the position changes, the processing unit 12 controls the position of the motorized platform 40 to correct the movement of the specimen so as to correct the defocus.

[0183] Preferably, the processing unit 12 further implements a trained Noise2Noise neural network to perform noise correction on each computed frontal image, more precisely, each computed frontal image is provided to the input of a neural network that outputs a denoised image.

[0184] Alternatively, as mentioned above, no modulation of the reference reflective surface 30 is performed. In this case, the processing unit 12 calculates the front image relying on the phase shift introduced from one interferogram to the other by the natural movement of the sample along the optical axis ZZ.

[0185] Second embodiment A second embodiment of an imaging device according to the present invention is shown in Figure 5. Reference numerals in common with Figure 1 correspond to the same elements.

[0186] The imaging device 2 of FIG. 5 is configured to determine at least one en face image, and preferably at least one cross sectional image, of the retina 7 of the in vivo eye 4.

[0187] To that end, the imaging device 2 of Figure 5 differs from the imaging device of Figure 1 in that the first lens system 34 does not consist of a microscope objective lens. In this case, the refractive elements of the eye focus an initial portion of the probe beam 20 onto the retina, thereby acting as the first lens system 34.

[0188] Preferably, the imaging device further includes an adaptive lens 76 (e.g., a liquid lens) and / or a rotating glass plate 78 inserted in the sample arm 26 to compensate for aberrations and dispersion mismatches introduced by the eye 4 with respect to the dispersion of the second lens system 36 in the reference arm 28.

[0189] If the optional conjugator 53 is provided, the conjugator 53 is positioned so that the first focal plane F1 of the conjugator 53 coincides with the corneal lens of the eye 4.

[0190] Furthermore, in such an imaging device, the processing unit 12 is advantageously configured such that, during operation, if the processing unit 12 determines that the sample 4 has moved (e.g., by determining that the cross-sectional segment structure of the sample has moved relative to a predetermined target center position), the processing unit 12 controls the motorized platform 40 so that the plane of the retina 7 at which the interference image (of the FF-OCT imaging system 8) occurs remains at the same position within the sample 4.

[0191] Third embodiment A third embodiment of an imaging device according to the present invention is shown in Figure 6. Numerical references in common with Figure 1 correspond to the same elements.

[0192] The imaging device 2 of Figure 6 differs from the imaging device of Figure 1 only in that the FF-OCT imaging system 8, positioning system 10 and sample beam splitter 19 are arranged so that the probe beam 20 is reflected by the sample beam splitter 19 and enters the sample 4, and the positioning beam 51 propagates through the sample beam splitter 19 and enters the sample 4.

[0193] In this case, the reference arm 28 does not include a reference beam splitter, since the probe beam 20 in the sample arm 26 does not undergo propagation through the sample beam splitter 19, which would cause dispersion that requires compensation.

[0194] Such a configuration can also be applied using the FF-OCT imaging system 8 of the imaging device 2 of FIG. 5, which is specially configured for retinal imaging.

[0195] Fourth embodiment A fourth embodiment of an imaging device according to the present invention is shown in Figure 7. Numerical references in common with Figure 1 correspond to the same elements.

[0196] The imaging device 2 of Figure 7 differs from the imaging device of Figure 1 only in that the positioning system 10 does not include an OCT imaging system but includes a stereoscopic imaging device 80 having an optical axis Δ as its optical axis, in which case positioning of the position of the sample 4 along the optical axis ZZ is achieved thanks to the depth determination capability provided by stereoscopic vision.

[0197] For example, as shown in FIG. 7A, a stereoscopic imaging device 80 is comprised of two cameras 82A, 82B coupled via a beam splitter 84, thereby providing a stereoscopic view. Alternatively, as shown in FIG. 7B, cameras 82A and 82B are placed on either side of the optical axis Δ, thereby achieving stereoscopic viewing.

[0198] Fifth embodiment A fifth embodiment of an imaging device according to the present invention is shown in Figure 8. Numerical references in common with Figure 1 correspond to the same elements.

[0199] In this embodiment, the position sensing system 10 includes an active position sensing device configured to emit an interrogation wave 90 along an axis Δ (or “sensing axis”). The interrogation wave is reflected by a sample beam splitter 19 and is incident on the sample 4.

[0200] The active position sensing is further configured to detect a response wave 92 that is backscattered by the sample 4, reflected by the sample beam splitter 19, and propagates along the detection axis Δ toward the position detection system 10. As a result, based on the detected response wave 92, the position detection system 10 outputs a position signal representative of the position of the sample 4 along the ZZ axis.

[0201] The interrogating wave may be light (eg, laser light) or an acoustic wave.

[0202] For example, the location system 10 is configured to determine a location signal based on time of flight.

Claims

1. An imaging device (2) for in vivo full-field interferometric imaging of a scattering sample (4), the imaging device (2) comprising: a full-field OCT imaging system (8) for providing at least one interferometric image representative of a front-on image of the sample, the full-field OCT imaging system (8) comprising: a primary light source (14) configured to output a probe beam (20) having low spatial coherence and a low temporal coherence length; a primary interferometer (16), a sample arm (26) for receiving the sample (4), a reference arm (28) provided with a reference reflective surface (30); a primary beam splitter (24) for separating said sample arm (26) and said reference arm (28); A primary interferometer (16) comprising at least: a camera (18) configured to acquire at least one interference image, Each interference image is representative of a first order optical interference signal resulting from interference occurring at each point of the imaging field when the sample (4) is placed in the sample arm (26), said interference being a reference wave obtained by reflection of an incident light wave of the probe beam (20) on a fundamental surface of the reference reflecting surface (30) conjugate to said point of the imaging field; a sample wave obtained by scattering an incident light wave of the probe beam (20) in an elementary volume of a slice of the sample (4) at a given depth, the elementary volume being conjugate to the point of the imaging field; and a camera (18) which occurs between With The imaging device (2) further comprises: a sample beam splitter (19) arranged in the sample arm (26) downstream of the primary beam splitter (24) in the direction from the primary light source (14) to the primary beam splitter (24); a sample localization system (10) arranged outside said sample arm (26), the sample localization system (10) having a respective detection axis (Δ) intersecting with a sample beam splitter (19), the sample localization system (10) being configured to output a position signal representative of the position of the sample (4) along the detection axis (Δ); An imaging device (2).

2. 2. The imaging device (2) according to claim 1, wherein the sample localization system (10) is an optical sample localization system, and the detection axis (.DELTA.) of the sample localization system (10) is the optical axis.

3. 3. The imaging device (2) according to claim 2, wherein the sample beam splitter (19) is a dichroic mirror.

4. 4. The imaging device (2) of claim 2 or 3, further comprising a reference beam splitter (33) arranged in the reference arm (28) between the primary beam splitter (24) and the reference reflective surface (30), the reference beam splitter (33) having the same optical properties as the sample beam splitter (19).

5. 5. The imaging device (2) of claim 4, wherein the reference beam splitter (33) is identical to the sample beam splitter (19), and the reference beam splitter (33) and the sample beam splitter (19) have a similar direction relative to the propagation direction of the light wave of the probe beam (20) propagating through each of them.

6. 6. The imaging device (2) of claim 2, wherein the optical sample localization system (10) comprises an OCT imaging system configured to determine, for each interference image acquired by the full-field OCT imaging system (8), at least one cross-sectional image of the sample (4) at the time of acquisition of the interference image.

7. The OCT imaging system (10), a secondary light source (46) configured to output a position beam (51); a second-order interferometer (48), a secondary sample arm (56) for receiving said sample (4), the optical axis of which is the optical axis of said optical sample positioning system (10); a secondary reference arm (58) consisting of a secondary reflector (66); a secondary beam splitter (54) for separating said secondary sample arm (56) and said secondary reference arm (58); a second-order interferometer (48) comprising: a detector (50) configured to output said position signal, said position signal being such that, when said sample (4) is placed in said secondary sample arm (56), said position signal a secondary reference wave obtained by reflection of the incident light wave of the position beam on a secondary reflector (66); a secondary sample wave resulting from scattering of the initial portion of the location beam by a volume of the sample (4) located along the propagation direction of the initial portion of the location beam (51); Detector (50), which is representative of the resulting secondary optical interference signal resulting from interference between: The imaging device (2) of claim 6, comprising:

8. The full-field OCT imaging system (8) includes a first lens system (34) disposed in the sample arm (26) downstream of the primary beam splitter (24) in a direction from the primary light source (14) to the primary beam splitter (24), the OCT imaging system further comprising: a beam scanner (70) including an optical deflection element (74) for deflecting a position beam (51), said beam scanner (70) being arranged in said secondary sample arm (56) between said sample beam splitter (19) and said secondary beam splitter (54); a conjugation device (53) comprising two lenses arranged in a 4F configuration, the rear focal plane of said first lens system (34) coinciding with a first focal plane (F1) of said conjugation device (53) and said optical deflection element (74) being arranged in a second focal plane (F2) of said conjugation device (53); The imaging device (2) of claim 7, comprising:

9. 9. The imaging device (2) according to any one of the preceding claims, further comprising at least one compensation lens (45) and / or a deformable lens and / or a variable focus lens (45) arranged in the sample arm (26) of the primary interferometer (16).

10. 10. The imaging device (2) according to any one of the preceding claims, further comprising a processing unit (12) configured to determine at least one front image of the sample (4) based on each acquired interference image and the corresponding position signal.

11. The processing unit (12) is further configured to perform a denoising on each front image, the denoising comprising: - acquiring at least one en face training image set using said full field OCT imaging system (8), where each en face training image set includes a first en face image and a second en face image at different acquisition times of the same section of the training sample (4); For each training set, provide the Noise2Noise model with the first frontal image as input and the second frontal image as target output.

11. The imaging device (2) of claim 10, further comprising providing the noise2noise neural network pre-trained by:

12. A method for in vivo full-field coherent imaging of a scattering sample (4), the method comprising the steps of: - placing a sample (4) on a sample arm (26) of a full-field OCT imaging system (8) according to any one of claims 1 to 11; - acquiring at least one position signal representative in time of the position of said sample (4) along a detection axis; - determining the position of the sample (4) over time based on each position signal; adjusting the position of the imaged slice of the sample based on the determined position of the sample.

13. The method of claim 12, wherein the sample (4) is an in vivo eye of a subject.

14. 14. The method of claim 13, wherein the sample arm (26) of the full-field OCT imaging system (8) includes a first lens system (34) configured to focus the probe beam (20) on the cornea (6) or retina (7) of the eye.

15. training a Noise2Noise neural network, - acquiring at least one en face training image set using said full field OCT imaging system (8), where each en face training image set includes a first en face image and a second en face image at different acquisition times of the same section of a training sample; for each training set, providing the first frontal image as an input to a Noise2Noise model and providing the second frontal image as a target output; moreover - determining at least one front image of the current sample (4) based on each acquired interference image and the corresponding position signal, Each determined frontal image is provided to the trained Noise2Noise neural network for noise removal.

15. The method of any one of claims 12 to 14, comprising:

Images