Full-field optical coherence tomography (FFOCT) imaging method with alignment adjustment

Abstract

A full-field optical coherence tomography (FFOCT) imaging method for acquiring an image facing a section of interest (15) in a sample (11), with a system comprising a first FFOCT imaging device performing simultaneous illumination of the sample arm (7) and the first reference arm (8) and acquiring an image facing the section of interest (15), and an alignment determination device (20) receiving a second sample light, with a second imager (32) configured to acquire a second image from a second reference light and the second sample light, the imaging method comprising, before the acquisition of the first two-dimensional image facing the section of interest (15), an acquisition of a second image by the second imager (32), representative of a cross-section of the sample, a cross-section (52) of the section of interest (15) appearing at a location of interest,and an alignment setting modifying at least one of the sample optical path and the reference optical path depending on the position of the location of interest. Figure for the abstract: FIG. 1,

Description

Title of the invention: Full-field optical coherence tomography (FFOCT) imaging method with alignment adjustment. Technical field

[0001] The present invention belongs to the field of optical coherence tomography, or OCT for the English "Optical Coherence Tomography" and more specifically relates to the adjustment of the coherence gate for the acquisition of an image in full-field optical coherence tomography, or FFOCT for the English "Full-Field Optical Coherence Tomography". State of the art

[0002] Full-field optical coherence tomography (FFOCT) is based on broadband interferometric microscopy. Tomographic images are obtained by combining interferometric images recorded by an imager such as a CCD or CMOS camera. While conventional OCT produces B-mode (axially oriented) images like ultrasound imaging, full-field OCT acquires tomographic images in the en face (transverse) orientation. More precisely, the interferometric images are created by an interferometer, where path length modulation is generally performed by an actuator (usually a mirror actuated by a piezoelectric element in the reference arm). During a modulation period, several images (usually 2 or 4) are acquired by an imager such as a CCD camera and then combined in post-processing (or online) using an algorithm based on the phase-shifting interferometry method.

[0003] The "en face" tomographic images are therefore produced by wide-field illumination. This can be achieved by the Linnik configuration of the interferometer, where a microscope objective is used in both arms. Furthermore, while the temporal coherence of the source must remain low, as in conventional OCT (i.e., a broad spectrum), the spatial coherence must also be low to avoid the overlaps that occur when a spatially coherent light source is used. Full-field OCT is an alternative method to conventional OCT for providing ultra-high-resolution images (~1 pm), using, for example, a simple halogen lamp instead of a complex source based on an ultrashort-pulse laser. Full-field OCT has several specific advantages.Because FFOCT acquires an image "front-on" without point-by-point or line-by-line transverse scanning, it is immune to transverse scanning artifacts. FFOCT offers higher lateral resolution (on the order of 1 pm) than conventional OCT (on the order of 10 pm) thanks to the use of large-aperture lenses. digital. This is particularly useful for examining microscopic cells and tissue structures in biological samples.

[0004] The full-field OCT imaging technique is described, for example, in the article "Full-field optical coherence tomography" by A. Dubois and C. Boccara, extracted from the book "Optical Coherence Tomography - Technology and Applications" - Wolfgang Drexler - James G. Fujimoto - Editors - Springer 2009. This is also disclosed in French patent application FR2817030.

[0005] FFOCT thus makes it possible to acquire a very high-quality image. However, it is desirable to maximize the spatial coincidence, that is, the superposition or overlap, between the coherence gate and the section of interest at a depth in the sample. Now, the section of interest may be curved.European patent application EP3839417 describes the use in at least one of the sample and reference arms of an optical curvature compensator which modifies a transverse variation distribution of an optical path length to compensate for the curved profile of the transverse variation distribution of the first optical path length, so that the transverse variation distribution of the reference optical path length traversed by the reference light incident on the imager and the transverse variation distribution of the second optical path length traversed by the light of interest incident on the imager coincide, resulting in the light of interest from the section of interest interfering with the reference light and the imager imaging the section of interest on a field of view of the imager to form the two-dimensional face FFOCT image acquired by the imager.

[0006] However, even when curving the coherence gate, it may not perfectly cover the section of interest. The quality, particularly the sufficient contrast, of the resulting image, and especially the surface area of ​​the good quality region in the resulting image, depends on the coverage of the section of interest by the coherence gate.

[0007] Fine adjustments are then necessary, requiring modification of the relative position between the sample and the FFOCT imaging device, or by moving the reference arm. This adjustment can be lengthy, taking several minutes, since it is then necessary to perform multiple image acquisitions until a satisfactory result is obtained. Adjusting the orientation of the coherence gate is therefore tedious and complicated, limiting the accessibility of the FFOCT and requiring a qualified operator. For ophthalmology in particular, it is important to minimize the duration of the examination to avoid generating additional fatigue for the patient and to maximize the number of usable images acquired during the examination. Indeed, the illumination of the eye can be unpleasant for the patient, and the translations or rotations of the patient's eye continually modify the orientation of the section of interest. This can delay the adjustment or necessitate a new adjustment. Consequently, image quality may be reduced (increased movement, difficulty focusing on the imager), and the number of images acquired may decrease. Furthermore, the adjustment process can be complex and may require the presence of a specialized technician in addition to medical personnel. Presentation of the invention

[0008] The invention aims to allow a simple and quick adjustment for an FFOCT acquisition, allowing to obtain quickly, intuitively, and easily a maximum recovery of the section of interest by the coherence gate.

[0009] To this end, a full-field optical coherence tomography (FFOCT) imaging method is proposed for acquiring a first two-dimensional FFOCT image of a section of interest in a sample at a depth along an imaging axis. The imaging method uses a system comprising a first FFOCT imaging device and the sample containing the section of interest to be imaged. The first FFOCT device comprises: - a first source of light, - a first imager, - a first beam splitter defining a sample arm and a first reference arm, the sample being placed at one end of the sample arm, the imaging process comprising: - simultaneous illumination at a first instant of illumination of the sample arm and the first reference arm with a first illumination light emitted by the first light source to generate a first sample light propagating along the imaging axis from the sample in the sample arm along a sample optical path and a first reference light propagating in the first reference arm from the first reflector along a reference optical path, - an acquisition of a first two-dimensional FFOCT image in front of the section of interest with the first imager from the first reference light and the first sample light combined in the beam splitter, the first sample light containing a first light of interest derived from the first illumination light and coming from the section of interest of the sample; The system includes an alignment determination device coupled to the first imaging device, the alignment determination device being configured to implement optical coherence tomography, and comprising an input arm configured to receive light from the first FFOCT imaging device, a second light source configured to emit a a second illumination light, a second reference arm connected to the second light source, a second reflector disposed at one end of the second reference arm, a coupler between the second reference arm and the input arm, and a second imager coupled to the coupler and configured to acquire a second image from a second reference light derived from the second illumination light, coming from the second reference arm, and from a second sample light derived from the second illumination light, coming from the sample arm, the imaging process further comprising, prior to the acquisition of the first two-dimensional frontal image of the section of interest: - simultaneous illumination at a second instant of illumination of the sample arm and the second reference arm with the second illumination light emitted by the second light source to generate the second sample light and the second reference light, - acquisition of a second image by the second imager from at least the second sample light and the second reference light combined in the coupler, said second image being representative of a cross-section of the sample along the imaging axis, a cross-section of the section of interest appearing at a location of interest in the second image, - an alignment setting modifying at least one of the sample optical path and the reference optical path depending on the position of the location of interest in the second image.

[0010] The invention is advantageously complemented by the following features, taken alone or in any technically possible combination thereof: - the alignment adjustment includes a modification of the first reference arm, a displacement of the first reference arm relative to the sample, or a tilting of an optical window in the first reference arm or in the sample arm; - the modification of the first reference arm includes a translation of a lens of the first reference arm and / or a first reflector of the first reference arm; -the method includes illuminating the first reference arm at the second instant of illumination by the second illumination light, to generate a coherence light propagating from the first reflector into the first reference arm, the input arm receiving the coherence light, the second image also being acquired from the coherence light and the second reference light combined in the coupler, a cross-section of a coherence gate appearing in the second image at a coherence location, and the alignment adjustment aims to make the coherence location coincide with the location of interest; - the system includes a memory storing positions of a plurality of locations of interest to which are assigned adjustment data values ​​corresponding to the modification of at least one of the sample optical path and the reference optical path as a function of the positions of each of said locations of interest in the second image, and the alignment adjustment includes the control of an actuator as a function of an adjustment data value corresponding to the position of the location of interest in the second image; - the adjustment data are obtained during a preliminary calibration phase during which the positions of the plurality of locations of interest are recorded for each of a plurality of adjustment data values; - the first light source emits in a first wavelength range, the second light source emits in a second wavelength range distinct from the first wavelength, and the first imaging device includes a spectral filter configured to block light derived from the second illumination light, the spectral filter being disposed in the first reference arm.

[0011] The invention relates to a full-field optical coherence tomography (FFOCT) imaging system for acquiring a first two-dimensional FFOCT image of a section of interest in a sample at a depth along an imaging axis. The imaging method uses a system comprising a first FFOCT imaging device and the sample comprising the section of interest to be imaged. The first FFOCT device comprises: - a first light source configured to emit an initial illumination light, - a first imager configured to acquire the first image, - a first beam splitter defining a sample arm and a first reference arm, the sample being disposed at one end of the sample arm, characterized in that the system further comprises an alignment determination device coupled to the first imaging device, the alignment determination device being configured to implement optical coherence tomography, and comprising an input arm configured to receive light from the first FFOCT imaging device, a second light source configured to emit a second illumination light, a second reference arm connected to the second light source, a second reflector disposed at one end of the second reference arm, a coupler between the second reference arm and the input arm,and a second imager coupled to the coupler and configured to acquire a second image from a second reference light derived from the second illumination light, coming from the second reference arm, and from a second, sample light derived from the second illumination light, originating from the sample arm, the system being configured to implement the method according to any one of the preceding claims.

[0012] The system may include an actuator configured to modify at least one of the sample optical path and the reference optical path. The actuator may be configured to modify the first reference arm, or to move the first reference arm relative to the sample, or to tilt an optical window in the first reference arm or in the sample arm. The system may further include a processing unit configured to generate a command for the actuator to implement the alignment adjustment based at least on the position of the location of interest in the second image. Presentation of the figures

[0013] The invention will be better understood from the following description, which relates to embodiments and variants of the present invention, given by way of non-limiting examples and explained with reference to the accompanying schematic drawings, in which:

[0014] - Figure 1 schematically shows an overview of the main components of the system according to the invention,

[0015] - [Fig.2] shows an example of a second image before alignment adjustment,

[0016] - [Fig.3] shows an example of the first image before the adjustment setting,

[0017] - [Fig.4] shows an example of a second image after alignment adjustment,

[0018] - [Fig.5] shows an example of the first image after the alignment adjustment. Detailed description

[0019] The term "coherence gate" refers to the path length of coincidence between the optical path lengths of a reference light and another light on which interference occurs. The concept of "coherence gate" is well-known in the field of FFOCT and takes on its usual meaning in that field. By extension, the coherence gate also designates the portion of the sample from which light originating from it will produce interference with a reference light.

[0020] For the purposes of non-limiting illustration, the first imaging device is hereinafter a full-field optical coherence tomography (FFOCT) imaging device. The invention can be used in the context of time-domain full-field OCT (FFOCT), sweep-source full-field OCT, Fourier-domain full-field OCT, or line-domain OCT time-domain, spectral-domain, Fourier-domain, or source-scanning OCT. It is sufficient that the first imaging device be capable of implementing OCT to acquire a first two-dimensional image that represents the sample along a surface intersected by the axis of propagation of the light rays in the sample.

[0021] For non-limiting illustrative purposes, the alignment determination device is described as implementing spectral-domain optical coherence tomography (SDOCT). The alignment determination device could also implement source-scanning OCT, online SD-OCT, or time-domain scanning OCT.The alignment determination device simply needs to be capable of implementing OCT to acquire a second two-dimensional cross-sectional image that represents the sample along a surface parallel to the axis of light propagation within the sample. The orientation of the imaged area in the second image by the alignment determination device 20 is therefore perpendicular to the imaged area in the first image by the first FFOCT imaging device.

[0022] Figure 1 shows an example of a full-field optical coherence tomography (FFOCT) device. A first light source 1, incoherent in space and time, such as a light-emitting diode (LED), emits a first light beam 2 at a first illumination instant, which is sent to a first beam splitter 3. The first light source 1 is spatially incoherent, meaning that it is a source where each point of the source emits waves whose phases are randomly distributed among those points. The first light source 1 must have a broad spectrum, typically between 20 nm and 150 nm wide (more preferably between 30 nm and 70 nm wide), between 700 nm and 900 nm, and more preferably around 750 nm and 870 nm. For example, the first light source 1 can be an LED or have a filament (like a halogen lamp).

[0023] The first beam splitter 3 divides the first incoming light beam 2 into a second light beam 5 and a third light beam 6. The second light beam 5 is sent to a sample arm 7 and the third light beam is sent to a first reference arm 8. The sample arm 7 and the first reference arm 8 are thus simultaneously illuminated by the same first illumination light. Typically, the setup is optically symmetrical and uses the same objective in both arms of the interferometer. The first reference arm 8 contains a microscope objective 10. In some applications, when the sample 11 is not an eye, the sample arm 7 may be equipped with a microscope objective having the same optical properties as the microscope objective 10 of the first reference arm 8. In the illustrated example, the method is applied to ocular imaging, and the sample 11 is an eye, which acts as a microscope objective for the light from the sample arm 7. It is therefore not necessary to equip the sample arm with a microscope objective.

[0024] In the first reference arm 8, the light from the third light beam 6 propagates along the optical reference path from the first beam splitter 3 to the reflector 12 and from the reflector 12 back to the first beam splitter 3. The second light beam 5 propagates along the optical sample path from the first beam splitter 3 to the end of the sample arm 7 facing the sample 11. The light exits the sample arm 7 and enters the sample IL

[0025] The end of the sample arm 7 focuses the second light beam 5 onto a part of the sample 11 (for example, as illustrated, a human eye cornea), collects the first sample light which is reflected at different depths in the sample 11, and transmits the first sample light to the first beam splitter 3. The microscope objective 10 of the first reference arm 8 focuses the third light beam 6 onto a first reflector 12, for example, a flat reference mirror, collects the first reference light which is reflected by the first reflector 12, and transmits the first reference light to the first beam splitter 3.

[0026] The first sample light contains light derived from the first illumination light emitted at the first instant of illumination, originating from different depths of the sample 11, and therefore from different layers of the sample 11. The first sample light from different layers of the sample 11 and the first reference light from the first reflector 12 recombine at the first beam splitter 3 and are focused by a tubular lens 13 onto a first imager 14, for example, a camera that acquires a two-dimensional image. The combination of the first sample light and the first reference light produces interference in the plane of the first imager 14, which captures it in a two-dimensional image opposite.

[0027] However, the interference is conditioned by a required trajectory correlation between the first sample light and the first reference light. This trajectory correlation along the field of view of the image can only be maintained for a portion of the first sample light that has a trajectory similar to that of the first reference light.

[0028] The first sample light is therefore made up of light components having different path lengths. The different path lengths cause the optical path length to vary depending on the depth from which the first light originates. sample light. The first reference light has traveled a reference path and therefore only interferes with the first sample light which has traveled a path equal to the length of the reference path in the thickness of the coherence gate, thus defining in the sample an interfering sample section which corresponds to the origins of the sample light which interferes with the first reference light.

[0029] The length of the reference path defined by the first reference arm 8 therefore defines the depth of the interfering sample section. Full-field OCT can thus be defined as an optical sectioning method, since it is capable of extracting the first sample light originating solely from the interfering sample section.

[0030] The thickness of the interfering sample section is determined by the spectral bandwidth of the first light source 1: the wider the spectrum of the first light source 1, the thinner the interfering sample section.

[0031] The first imager 14, however, collects all the first sample light from the first beam splitter 3, that is, the interfering light from the interfering sample section and the non-interfering light from the rest of the sample 11 (sections located in front of or behind the interfering sample section along the optical axis Z). Such an acquired image appears blurred, containing the superposition of the interfering sample section and other sections of the sample near the interfering sample section. It is therefore necessary to eliminate the non-interfering light from the rest of the sample.

[0032] This is generally achieved by acquiring a series of several images (typically 2 to 5 images) of the same sample 11, with a modulated interference phase. The first reflector 12, for example a reference mirror, can be translated using a piezoelectric element that generates an oscillation of the position of the first reflector 12, thus modulating the length of the reference path and therefore the interference phase. Each acquired image corresponds to a particular interference phase. Post-processing of the acquired image series eliminates the non-interfering light, and the resulting full-field OCT image reveals only the interfering light from the sample originating from the portion of interest of the sample 11. This yields a final two-dimensional (2D) image of an interesting section of the sample 11.

[0033] The optical path length is the product of the geometric length of the path followed by the light and the average refractive index of the medium in which it propagates. A light beam does not propagate solely along the optical axis, due to its spatial extension in a transverse plane perpendicular to the optical axis, and the optical path length can vary depending on the position considered on said axis. transverse plane. This is what is called the distribution of the transverse variation of the optical path length. The distribution of transverse variations of the optical path length has a profile defined by the variations of the optical path length on the transverse plane.

[0034] If the optical path length is the same for all points in the transverse plane, the optical path length profile is planar. If the optical path length varies depending on the position considered on said transverse plane, the optical path length profile is not planar. For example, if peripheral points have a longer or shorter optical path than points located at the center (i.e., on the optical axis), the optical path length profile can be defined as curved.

[0035] The imaging method aims to acquire an image of a specific section of interest 15. The section of interest 15 is defined by a depth within the sample 11 and by a shape. Generally, the section of interest 15 is not flat but rather has a curved surface relative to the optical axis of the FFOCT device. For example, when the sample 11 is an eye, a section of interest 15 may have a convex shape as viewed from the sample arm 7, for example, for a comelina layer or a lens, or a concave shape as viewed from the sample arm 7, for example, for a retinal layer.

[0036] Preferably, the FFOCT device is equipped with an optical curvature compensator that modifies the distribution of the transverse variation in the optical path length. The optical curvature compensator can be placed in the sample arm 7 or in the first reference arm 8. The optical curvature compensator is configured to compensate for the relative curvatures of the profiles of the first reference light and the light of interest. When the optical curvature compensator is placed in the first reference arm 8, it modifies the reference optical length in the image field of view. When the optical curvature compensator is placed in the sample arm 7, it modifies the optical length of the sample 11 in the imager field of view 14.The optical curvature compensator is configured so that the length of the reference optical path traveled by the first reference light incident on the first imager 14 or the second length of the optical path traveled by the light of interest incident on the first imager 14 have the same transverse variation distribution, i.e. that the transverse variation distribution of the length of the reference optical path traveled by the first reference light incident on the first imager 14 and the transverse variation distribution of the second length of the optical path traveled by the light of interest incident on the first imager 14 coincide within the temporal coherence length.

[0037] Preferably, the profiles of the two respective transverse variation distributions have an absolute radius of curvature difference of less than 2 millimeters, and more preferably less than 1 mm. When the light reaches the imager 14, the light waves of the light of interest and the first reference light interfere and produce interference. The interference occurs at each point in the field of view of the imager 14 when the differences in optical path lengths of the waves overlap in the temporal coherence length defined by the first illumination light. The proposed optical curvature compensation thus allows the light of interest from the section of interest 15 to interfere with the first reference light in the plane of the imager 14, and allows the imager 14 to image the section of interest over a large part of its field of view, and possibly its entire field of view.

[0038] The optical curvature compensator can be a reflector with a curved reflective surface disposed in the first reference arm 8. The optical curvature compensator can therefore be the first reflector 12 disposed at one end of the reference arm 8 opposite the beam splitter 3, if the first reflector 12 is curved. In [Fig. 1], the section of interest 15 is an anterior surface of the ocular lens, which is convex as seen from the FFOCT device. The first reflector 12 is also curved into a convex shape as seen from the optical reference path. The curvature of the first reflector 12 corresponds to the curvature of the anterior lens, and the radius of curvature is approximately between 9 and 15 mm, and more preferably between 11 and 13 mm. If the section of interest 15 is the anterior cornea, the radius of curvature of the reflector 12 is between 7 mm and 8 mm.If the section of interest 15 is the posterior cornea, the radius of curvature of reflector 12 is between 6 mm and 7 mm. The first reflector 12 can also be curved into a concave shape as viewed from the reference optical path. The curvature of the first reflector 12 then corresponds to the curvature of the retinal layer, and the radius of curvature is approximately between -11 mm and -13 mm. The microscope objective in the sample arm 7 can be omitted, as in ocular imaging such as the illustrated example. Removing the microscope objective from the sample arm 7 results in a strong asymmetry between the two arms 7 and 8 of the FFOCT device. This asymmetry, combined with the asymmetry caused by the ocular medium opposite the retinal layer of interest 15, leads the second optical path length traveled by the light of interest incident on the first imager 14 to a transverse variation distribution with a curved profile.The curvature of the first reflector 12 can therefore be chosen to be greater than the curvature of the retinal layer of interest, in order to compensate for the asymmetry.

[0039] The curved shape of the first curved reflector 12 causes the reflected light to travel an optical path length that varies through the cross-section of the reference optical path, thus causing a transverse variation distribution of the reference optical path length. By choosing the curvature of the first curved reflector 12 to match the curvature of the section of interest 15, the length of the reference optical path traveled by the first reference light incident on the imager 14 and the length of the optical path traveled by the light of interest incident on the imager 14 have the same transverse variation distribution. Therefore, the optical sectioning performed by the first reference light on the first sample light selects only the light of interest.

[0040] The first curved reflector 12 can be a mirror, and in particular a curved metallic mirror, for example with an aluminum coating as its reflective surface. The first curved reflector 12 is not necessarily a mirror and can, for example, be an optical lens. Such an optical lens can be made of glass (for example, RoHS-compliant borosilicate glass), fused silica, or any other suitable material. The optical curvature compensator can also be a plate of material having a compensating refractive index and a compensating length along the reference optical path or the sample optical path. Such an optical curvature compensator can therefore be placed in the first reference arm 8 or in the sample arm 7. The plate, or optical window, is an optically flat piece of transparent optical material.

[0041] Further details can be found in patent application WO2021 / 123257 on the full-field optical coherence tomography imaging method, FFOCT, with an optical compensator modifying a transverse variation distribution of an optical path length to compensate for the curved profile of the transverse variation distribution of the first optical path length, so that the transverse variation distribution of the reference optical path length traversed by the first reference light incident on the imager and the transverse variation distribution of the second optical path length traversed by the light of interest incident on the imager coincide, which allows the light of interest from the section of interest to interfere with the first reference light and the first imager 14 to image the section of interest over a large part of its field of view,and possibly its entire field of view, to form the two-dimensional FFOCT image acquired by the first imager 14.

[0042] As mentioned above, even when curving the coherence gate, it may not perfectly coincide with the section of interest. The quality of the resulting image (essentially the contrast), and especially the surface extent of the area of ​​good quality in the resulting two-dimensional image, depends on the overlap of the section of interest by the coherence gate.

[0043] To this end, the system includes an alignment determination device 20 coupled to the first FFOCT imaging device. The alignment determination device 20 is configured to acquire a second cross-sectional image, rather than a frontal one, in particular by implementing optical coherence tomography, specifically in the spectral domain (SDOCT). The alignment determination device 20 allows the adjustment of the first imaging device to align the section of interest 15 with the coherence gate that enables its acquisition.

[0044] The alignment determination device 20 comprises an input arm 22 configured to receive light from the first FFOCT imaging device, a second light source 24 configured to emit a second illumination light, a second reference arm 26 connected to the second light source 24, a second reflector 28 disposed at one end of the second reference arm 26, a coupler 30 between the second reference arm 26 and the input arm 22, and a second imager 32 coupled to the coupler 30 and configured to acquire a second image. Typically, the various elements of the alignment determination device 20 are connected by a single-mode fiber 34, one portion of which forms the second reference arm 26 and another portion forms the input arm 22.

[0045] The second light source 24 is preferably incoherent in space and time, such as a light-emitting diode, or LED, or a superluminescent diode, or SLD, and is configured to emit a second illumination light transmitted through the optical fiber 34 to the coupler 30. The second light source 24 is preferably spatially incoherent, that is, it is a source where each point of the source emits waves whose phases are randomly distributed among those points. The second light source 24 emits in a second wavelength range that may be the same as, different from, or distinct from the first wavelength range in which the first light source 1 emits.The second light source 24 typically has an emission spectrum with a width between 20 nm and 150 nm (more preferably between 30 nm and 70 nm), preferably between 870 nm and 1000 nm, and more preferably between 900 nm and 950 nm. The second reflector 28 at the end of the second reference arm 26 is typically a plane mirror.

[0046] The second imager 32, like the first imager 14, can be any suitable device for acquiring an image, with the appropriate qualities for this purpose (sensitivity to received light, speed, etc.). The imager 14, 32 can, for example, be a CCD or CMOS sensor. The second imager 32 can, however, be simpler than the first imager 14, and the second image less detailed than the first image, since the second image is essentially used to determine a location of interest within it. The second imager 32 could, for example, be one-dimensional.

[0047] The input arm 22 is optically coupled via an interface 23 to a scanner 36, for example, a galvanometer scanner, allowing optional scanning of the field of view of the alignment determination device 20. In this example, the scanner 36 comprises two mirrors, at least one of which is movable, and preferably both mirrors are movable. The scanner 36 is optically coupled to a second beam splitter 38, for example, a semi-reflective plate or a dichroic mirror. The second beam splitter 38 establishes optical communication between the scanner 36 and the first beam splitter 3. The second beam splitter 38 can be configured to receive the first illumination light from the first light source 1 and redirect it to the first beam splitter 3.

[0048] The alignment determination device 20 is coupled to the first FFOCT imaging device via the input arm 22. The second illumination light thus enters the first FFOCT imaging device via the input arm 22 and propagates into the sample arm 7 and the first reference arm 8. The second illumination light thus generates a second sample light propagating along the imaging axis 16 from the sample 11 into the sample arm 7 along a sample optical path, and a coherence light propagating from the first reflector 12 into the first reference arm 8. The second illumination light also propagates into the second reference arm 26 and generates a second reference light propagating from the second reflector 28 into the second reference arm 26.

[0049] The second beam splitter 38 is configured to receive from the first beam splitter 3 the second sample light which contains a light of interest derived from the second illumination light and coming from the section of interest 15 of the sample 11, and to transmit this light of interest to the interface 23 of the input arm 22 of the alignment determination device 20. The coherence light also passes from the first beam splitter 3 to the interface 23 of the input arm 22 of the alignment determination device 20.

[0050] In this example, the input arm 22 of the alignment determination device 20 is positioned on the opposite side of the first reference arm 8 from the first beam splitter 3. This arrangement is not mandatory, and the input arm 22 could be positioned elsewhere as long as it can receive the second sample light and possibly the coherence light. For example, the alignment determination device 20 could be positioned in parallel with the first light source 1, or even in parallel with the first imager 14. It is also possible to exchange the position of the alignment device 20 and the light source 1 relative to the first beam splitter 3.

[0051] Preferably, the first imaging device includes an actuator 40 configured to modify at least one of the sample optical path and the reference optical path. The actuator 40 allows for alignment adjustment. The actuator 40 can be configured to modify the first reference arm 8, move the first reference arm 8 relative to the sample 11, or tilt an optical window 42 positioned in the first reference arm 8 or in the sample arm 7, for example, between the beam splitter 3 and the sample 11. For example, as in [Fig. 1], modifying the first reference arm 8 can include translating the lens 10 of the first reference arm 8 and / or the first reflector 12 of the first reference arm 8. Preferably, the actuator 40 allows translation along two axes x, y perpendicular to the z-axis of light propagation.For example, the reference arm 8 can be mounted on two motors whose positions are known and precisely controlled along the x and y axes, respectively. The actuator 40 can thus comprise a motor, a motorized screw, a cylinder, or any other element enabling alignment adjustment.

[0052] The alignment determination device 20 allows for the implementation of an alignment adjustment modifying at least one of the sample optical path and the reference optical path in order to make the section of interest 15 coincide with the coherence gate allowing the imaging of said section of interest 15.

[0053] Thus, the imaging method comprises, in addition to the simultaneous illumination of the sample arm 7 and the first reference arm 8 with a first illumination light emitted by the first light source 1 at a first illumination instant, the emission of a second illumination light by the second light source 24 at a second illumination instant. This second illumination light, by illuminating the sample arm 7, generates the second sample light, which, similarly to the first sample light, contains light derived from the second illumination light emitted at a second illumination instant, originating from different depths of the sample 11, and therefore from different layers of the sample 11.

[0054] The second sample light enters the alignment determination device 20 through the inlet arm 22, up to the coupler 30.

[0055] The second illumination light propagates in the second reference arm 26, is reflected by the second reflector 28 and becomes a second reference light, which travels back through the second reference arm 26 to the coupler 30, where it combines with the second sample light.

[0056] As with the first sample light, the second sample light consists of light components with different path lengths. These different path lengths cause the optical path length to vary depending on the depth from which the second sample light originates. The second reference light has traveled a second reference path and therefore only interferes with the second sample light, which has traveled a path equal to the length of the second reference path, up to the coherence length. This defines an interfering sample section in sample 11, corresponding to the origins of the second sample light that interferes with the second reference light. Fiber 34 carries the combination of the second sample light and the second reference light from coupler 30 to the second imager 32.

[0057] The second illumination light also illuminates the first reference arm 8, and generates a coherence light propagating from the first reflector 12 into the first reference arm 8. The coherence light passes through the first beam splitter 3, and the input arm 22 receives the coherence light, in this example via the second beam splitter 38 and the scanner 36.

[0058] The second imager 32 acquires a second two-dimensional image from at least the second sample light and the second reference light combined in the coupler 30. The second image represents a cross-section of the sample 11 along the imaging axis, i.e., along the z-axis of light propagation. As illustrated in the example in [Fig. 2], which shows a second image 50, a cross-section 52 of the section of interest 15 appears at a location of interest in the second image 50. This representation of the cross-section 52 of the section of interest 15 results from the light interference between the second sample light and the second reference light.

[0059] In this second image, at a position designated as the coherence location, there also appears a representation of a cross-section 54 of a coherence gate resulting from the interference between the coherence light and the second reference light, combined in the coupler 30. It can be seen here that the location of interest and the coherence location do not coincide, although the section of interest 15 and the coherence gate have the same curvature. It follows that a first FFOCT image acquired by the first imaging device can correctly represent only a portion of the section of interest 15 that corresponds to the intersection with the coherence gate, as illustrated in the example in [Fig. 3].

[0060] An alignment adjustment is therefore performed, modifying at least one of the sample optical path and the reference optical path according to the position of the location of interest in the second image. This alignment adjustment can be manual, by modifying the first reference arm 8, by moving the first imaging device with respect to sample 11, or by moving the optical window 42 positioned between the first beam splitter 3 and sample 11. A display screen is then used to allow the operator to view the second image.

[0061] Preferably, it is the first reference arm 8 that is modified, by a translation of the lens 10 of the first reference arm 8 and / or the first reflector 12 of the first reference arm 8, for example in the x and / or y directions perpendicular to the propagation direction z. This manual adjustment can be guided by the respective representations of section 52 of the section of interest 15 and section 54 of the coherence gate: the aim is to make the coherence location coincide with the location of interest in the second image. The alignment adjustment moves the coherence location, and when it best overlaps the location of interest, the alignment adjustment is complete.

[0062] Figure 4 shows an example of a second image after the alignment adjustment has been made. The coherence location where slice 54 of the coherence gate is located is superimposed on the location of interest where slice 52 of the section of interest 15 is located. This means that the first imaging device can correctly image a larger area of ​​the section of interest 15. Figure 5 shows an example of the corresponding first image. Comparing it with the first image of Figure 3, it can be seen that, thanks to the alignment adjustment, a larger area of ​​the section of interest 15 appears in the first image, filling the entire field of view of the imager 14 of the first imaging device.

[0063] Preferably, however, the alignment adjustment is not performed manually by an operator observing the position of the location of interest in the second image, but is performed automatically. To this end, in addition to a processor, the system includes a memory storing the positions of a plurality of locations of interest to which are assigned adjustment data values ​​corresponding to the modification of at least one of the sample optical path and the reference optical path as a function of the positions of each of said locations of interest in the second image.

[0064] Thus, following the acquisition of a second image by the second imager, the processor is configured to detect in the second image the location of interest where section 52 of the section of interest 15 is located, for example by segmentation or edge detection. The location of interest can also be entered by an operator viewing the second image on a display screen.

[0065] Once the location of interest has been determined in the second image, a match is sought in memory with the positions of a plurality of locations of interest. Conventionally, this can be done by selecting the nearest stored location of interest, or by interpolating between positions of the nearest locations of interest, or any other method. Memory can also store this information as a simple relationship, for example an equation, matching positions of a plurality of locations of interest to tuning data values.

[0066] In all cases, adjustment data values ​​are obtained, the application of which to the actuator 40 enables the alignment adjustment. For example, if it is determined that the slice 52 of the section of interest 15 in the second image corresponds to adjustment data values ​​xl, yl, then the actuator is controlled according to commands enabling the movement of the lens 10 of the first reference arm and / or the first reflector 12 of the first reference arm 8 to a position where it has been determined that the section of interest and the coherence gate used to image it will coincide. This adjustment data is obtained during a preliminary calibration phase in which the positions of the plurality of locations of interest are recorded for each of a plurality of adjustment data values, or vice versa.

[0067] When using adjustment data, it is not necessary to have the coherence gate slice 54. It is therefore preferable to remove the representation of this slice 54 from the second image. To do this, it is preferable to eliminate the second reference light from the alignment determination device 20. For this purpose, a spectral filter SPF can be placed in the first imaging device, preferably in the first reference arm as in the illustrated example. The spectral filter SPF is chosen to allow light waves to pass through in the first wavelength range and to block light waves in the second wavelength range. The filter is thus configured to block light derived from the second illumination light, such as the coherence light.

[0068] It is also possible to use the alignment determination device 20 and the second images acquired by it to implement position tracking of the section of interest 15 of the sample 11, based on the representation of the slice 52 of the section of interest 15 in the second image. By detecting the location of interest in the second image, the processor can control the actuator 40, or any other means of moving the first illumination light and the imaged area. For example, if the sample 11 is an eye, it becomes possible to track the movements of the retina. To facilitate this tracking, it is preferable to eliminate the coherence light from the first reference arm 8, as mentioned above, so that there is no other element in the second image than the representation of the slice 52 of the section of interest 15.

[0069] Preferably, the system includes a processing unit configured to generate a command for the actuator 40 to implement the alignment adjustment based at least on the position of the location of interest in the second image. The processing unit typically includes a processor and memory, and may have access to the adjustment data. The processing unit is, for example, a computer.

[0070] Of course, once the alignment adjustment has been made, all that remains is to acquire the first two-dimensional image opposite the section of interest 15 with the first imager 14.

[0071] Thanks to the alignment determination device 20, it is therefore possible to substantially accelerate the adjustment of the first imaging device, and thus the subsequent acquisition of the first image, while allowing the entire field of view of the first imaging device to be exploited.

[0072] It should be noted that the alignment adjustment can also be used to adjust the respective curvatures of the cut 52 of the section of interest 15 and of the cut 54 of the coherence gate, by adjusting an optical curvature compensator which modifies the distribution of the transverse variation of the optical path length.

[0073] The invention is not limited to the embodiment described and shown in the accompanying figures. Modifications remain possible, particularly with regard to the composition of the various elements or by substitution of technical equivalents, without departing from the scope of protection of the invention.

Claims

1. Demands A full-field optical coherence tomography (FFOCT) imaging method for acquiring a first two-dimensional FFOCT image of a section of interest (15) in a sample (11) at a depth along an imaging axis (16), the imaging method using a system comprising a first FFOCT imaging device and the sample comprising the section of interest (15) to be imaged, the first FFOCT device comprising: - a first light source (1), - a first imager (14), - a first beam splitter (3) defining a sample arm (7) and a first reference arm (8), the sample (11) being disposed at one end of the sample arm (7), the imaging method comprising: - simultaneous illumination at a first instant of illumination of the sample arm (7) and the first reference arm (8) with a first illumination light emitted by the first light source (1) to generate a first sample light propagating along the imaging axis (16) from the sample (11) in the sample arm (7) along a sample optical path and a first reference light propagating in the first reference arm (8) from the first reflector (12) along a reference optical path, - an acquisition of a first two-dimensional FFOCT image in front of the section of interest (15) with the first imager (14) from the first reference light and the first sample light combined in the beam splitter (3), the first sample light containing a first light of interest derived from the first illumination light and coming from the section of interest of the sample (11); characterized in that the system comprises an alignment determination device (20) coupled to the first imaging device, the alignment determination device (20) being configured to implement optical coherence tomography, and comprising an input arm (22) configured to receive light from the first FFOCT imaging device, a second light source (24) configured to emit a a second illumination light, a second reference arm (26) connected to the second light source (24), a second reflector (28) disposed at one end of the second reference arm (26), a coupler (30) between the second reference arm (26) and the input arm (22), and a second imager (32) coupled to the coupler (30) and configured to acquire a second image from a second reference light derived from the second illumination light, coming from the second reference arm (26), and from a second sample light derived from the second illumination light, coming from the sample arm (7), the imaging method further comprising, before the acquisition of the first two-dimensional face image of the section of interest (15): - simultaneous illumination at a second illumination instant of the sample arm (7),and the second reference arm (26) with the second illumination light emitted by the second light source (24) to generate the second sample light, and the second reference light, - an acquisition of a second image by the second imager (32) from at least the second sample light and the second reference light combined in the coupler (30), said second image being representative of a cross-section of the sample (11) along the imaging axis, a cross-section (52) of the section of interest (15) appearing at a location of interest in the second image, - an alignment adjustment modifying at least one of the sample optical path and the reference optical path as a function of the position of the location of interest in the second image, in which the alignment adjustment includes a modification of the first reference arm (8), a displacement of the first reference arm (8) relative to the sample (11), or a tilting of an optical window (42) in the first reference arm (8) or in the sample arm (7), the modification of the first reference arm (8) comprising a translation of a lens (10) of the first reference arm (8) and / or of a first reflector (12) of the first reference arm (8) along two axes perpendicular to the axis of light propagation. A method according to the preceding claim, comprising illuminating the first reference arm (8) at the second illumination instant with the second illumination light, to generate a light of coherence propagating from the first reflector (12) into the first reference arm (8), the input arm (22) receiving the coherence light, the second image also being acquired from the coherence light and the second reference light combined in the coupler (30), a slice (54) of a coherence gate appearing in the second image at a coherence location, and the alignment adjustment is intended to make the coherence location coincide with the location of interest.

3. A method according to any one of the preceding claims, wherein the system comprises a memory storing positions of a plurality of locations of interest to which are assigned adjustment data values ​​corresponding to the modification of at least one of the sample optical path and the reference optical path as a function of the positions of each of said locations of interest in the second image, and the alignment adjustment comprises the control of an actuator (40) as a function of an adjustment data value corresponding to the position of the location of interest in the second image.

4. A method according to the preceding claim, wherein the adjustment data are obtained during a preliminary calibration phase in which the positions of the plurality of locations of interest are recorded for each of a plurality of adjustment data values.

5. A method according to any one of the preceding claims, wherein the first light source (1) emits in a first wavelength range, the second light source (24) emits in a second wavelength range distinct from the first wavelength, and the first imaging device includes a spectral filter (SPF) configured to block light derived from the second illumination light, the spectral filter being disposed in the first reference arm (8).

6. A full-field optical coherence tomography (FFOCT) imaging system for acquiring a first two-dimensional FFOCT image of a section of interest (15) in a sample (11) at a depth along an imaging axis (16), the imaging method using a system comprising a first device

7. FFOCT imaging and the sample comprising the section of interest (15) to be imaged, the first FFOCT device comprising: - a first light source (1) configured to emit a first illumination light, - a first imager (14) configured to acquire the first image, - a first beam splitter (3) defining a sample arm (7) and a first reference arm (8), the sample (11) being disposed at one end of the sample arm (7), characterized in that the system further comprises an alignment determination device (20) coupled to the first imaging device, the alignment determination device (20) being configured to implement optical coherence tomography, and comprising an input arm (22) configured to receive light from the first FFOCT imaging device, a second light source (24) configured to emit a second illumination light, a second reference arm (26) connected to the second light source (24), a second reflector (28) disposed at one end of the second reference arm (26), a coupler (30) between the second reference arm (26) and the input arm (22),and a second imager (32) coupled to the coupler (30) and configured to acquire a second image from a second reference light derived from the second illumination light, originating from the second reference arm (26), and from a second sample light derived from the second illumination light, originating from the sample arm (7), the system being configured to implement the method according to any one of the preceding claims, the system further comprising an actuator (40) configured to modify at least one of the sample optical path and the reference optical path, along two axes perpendicular to the light propagation axis, by modifying the first reference arm (8), or to move the first reference arm (8) relative to the sample (11), or to tilt an optical window (42) in the first reference arm (8) or in the sample arm (7). System according to the preceding claim, further comprising a processing unit configured to generate a command for the actuator (40) to implement the alignment adjustment in function at least of the position of the location of interest in the second image.

Images