Device and method for interferometric imaging of a diffusing object

Abstract

The invention relates to an interferometric imaging device (1) comprising: - an illumination module (2) for generating a measurement beam (10) producing a line of light on an object as well as a reference beam (11); - interferometric means (8, 9) for generating an interference signal between the backscattered measurement beam and the reference beam; - a two-dimensional sensor (14) comprising a rolling shutter; - scanning means for moving the line of light on the object; - a synchronization module having a first mode with a zero time shift between the rolling shutter and the scanning means, the sensor sensing the interference signal, and a second mode with a non-zero time shift between the rolling shutter and the scanning means, the sensor sensing an incoherent signal; - a module for producing images from the interference signals and the incoherent signals.

Description

Device and method for interferometric imaging of a scattering object

[0001] The present invention relates to a device for interferometric imaging of a scattering object. It also relates to a method for imaging such an object.

[0002] The field of the invention is, in a non-limiting manner, that of imaging of diffusing media or objects, such as biological tissues. State of the art

[0003] Many imaging techniques are implemented in all scientific fields, particularly in medicine.

[0004] For imaging complex biological tissues, and particularly the eye in vivo, we can distinguish between incoherent and coherent techniques.

[0005] Incoherent techniques can be based on direct reflectance imaging or on phase and absorption imaging.

[0006] Coherent techniques are based on interferometric measurements of light backscattered by certain tissue structures.

[0007] Incoherent imaging techniques are generally based on direct measurement of reflectance.

[0008] In incoherent imaging, a full-field configuration allows the measurement of photons backscattered by the object with a camera. This is a fast technique allowing all photons to be recorded in parallel. However, it is not very sensitive because not only photons backscattered by a layer of interest are captured, but the camera also detects photons scattered by other layers or multi-scattered photons, which results in a high background noise.

[0009] In confocal imaging techniques, measurements are performed point by point with a scanning system in both transverse dimensions to obtain a two-dimensional image. Confocal filtering is performed to filter out photons from depths other than that of the measured object and multiply scattered photons, and therefore allow imaging with better contrast. However, this technique suffers from a low acquisition speed due to scanning.

[0010] The axial resolution of these techniques depends on the numerical aperture of the measuring system.

[0011] Coherent imaging techniques are generally based on the interferometric measurement of backscattered light. Coherent techniques provide access to the field and phase of the captured signals.

[0012] An important coherent technique is optical coherence tomography (OCT). Optical coherence tomography is an optical microscopy method for imaging, among other things, complex scattering biological tissues by selecting an imaging depth using low-coherence interferometry. In this interferometric technique, an interference signal between the light scattered by a sample and a reference beam is detected. In low coherence, for example in white light, the interference is localized to a narrow area defining the axial resolution of the measuring system. The axial resolution is thus decoupled from the depth of field and can therefore be arbitrarily small. Thanks to interferometry, small signals are amplified, making OCT particularly sensitive.

[0013] Among the OCT techniques, we distinguish Fourier domain OCT (FD-OCT) and time domain OCT.

[0014] In FD-OCT, the spectrum of the interferometric signal coming from a column of the sample (and therefore from different depths) is recorded, and the axial information of the scatterers is recovered by digital processing (Fourier Transform). We distinguish between spectral domain OCT, which measures the spectrum using a diffraction grating and a linear detector, and source scanning OCT, which measures the spectrum by successive measurements via a point detector by scanning the wavelength of the optical source. The amplitude and phase of the sample beam are obtained directly by the Fourier Transform operation.

[0015] In TD-OCT, the interference signal from a given depth is measured directly by integrating all wavelengths via a point detector. The depth information can be obtained by performing an axial scan of the reference mirror. The amplitude and phase of the incident signal can be recovered by performing, for example, a so-called 4-phase modulation, which consists of taking 4 successive images by shifting the reference arm by pi / 2 between each image. By cleverly combining the images, it is possible to filter out the photons coming from the coherence plane alone, and to separate the interference amplitude term and the phase shift term.

[0016] A distinction is also made between scanning OCT, in which measurements in the transverse plane are made point by point by scanning, and full-field OCT, in which an interference signal is measured in a two-dimensional transverse plane and recorded by a camera.

[0017] In scanning TD-OCT, an axial scan of the reference arm as well as a transverse scan of the scanned positions are required to image an entire volume. In scanning FD-OCT, at each point in the field of view, information of the entire depth of the sample is obtained via a one-dimensional measurement, typically with a spectrometer, but a two-dimensional scan is required.

[0018] In contrast, full-field OCT requires scanning only in the axial dimension of the object or sample, and therefore allows high imaging rates. However, this technique has a relatively low sensitivity, because the photons that do not interfere are captured by the detector, and although filtered, they contribute to creating noise (shot noise) on the detector. It may therefore be advantageous to filter these photons otherwise, for example with a confocal filter, which is only possible in scanning OCT. In full-field OCT, the acquisition of all image pixels simultaneously prevents the use of confocal filtering. The use of a confocal filter is also important to limit the influence of parasitic interference (cross-talk) due to multi-scattered photons.

[0019] The invention aims to resolve the disadvantages of the prior art described.

[0020] It is in particular an object of the invention to propose a device and a method for interferometric imaging of a scattering object making it possible to carry out full-field imaging at high speed and sensitivity.

[0021] It is another object of the present invention to provide a device and a method for interferometric imaging of a scattering object allowing both interferometric imaging of a scattering object and incoherent dark field imaging, in order to be able to advantageously combine different contrasts.

[0022] It is also an object of the present invention to propose a device for interferometric imaging of a diffusing object whose configuration is simple and compact.

[0023] At least one of these aims is achieved with a device for interferometric imaging of a scattering object, comprising:a low coherence linear illumination module configured to generate a measurement beam and a reference beam of linear section, the measurement beam producing a light line-shaped illumination of the object in a field of view;interferometric means configured to generate an interference signal resulting from interference between the measurement beam backscattered in the light line and the reference beam;a two-dimensional sensor comprising a rolling shutter;scanning means configured to move the light line over the object in the field of view;a synchronization module configured to operate in a first synchronized mode with a zero time offset between the rolling shutter and the scanning means, in which the sensor is configured to capture the interference signal, and in a second synchronized mode with a determined non-zero time offset between the rolling shutter and the scanning means, in which the sensor is configured to capture a light signal, called incoherent, coming from a non-illuminated area of ​​the object; andan image processing module configured to produce a bright field image from the interference signals and to produce a dark field image from the incoherent signals.;

[0024] The imaging device according to the present invention makes it possible to combine the advantages of a full-field configuration of an imaging device based on low-coherence interferometry and a dark-field configuration, and to combine the resulting contrasts. The rolling shutter of the sensor, and thus the successive exposure of lines of pixels of the sensor, acts as a real-time spatial filtering, making it possible to considerably increase the detection sensitivity compared to a full-field configuration without filtering, and without the need for additional digital filtering during image processing.

[0025] The device according to the invention makes it possible to carry out interferometric imaging in bright field and dark field, the two configurations being independent of each other. The device according to the invention makes it possible to produce images only in bright field or only in dark field, or both.

[0026] The device according to the invention thus makes it possible to perform bright-field interferometric imaging. In the configuration of the first synchronization mode, the illumination is synchronized with the rolling shutter (or dynamic digital diaphragm) so that only the pixels exposed at each time step are illuminated by light backscattered by the object and by the reference arm. It is possible to perform low-coherence interferometry to image a scattering object and obtain reflectivity contrast images. In particular, bright-field imaging makes it possible to obtain images of scattering structures with very good contrast. The imaging speed in this configuration is the same as in a full-field configuration, while increasing the detection sensitivity thanks to spatial filtering.

[0027] The image processing module is configured to produce a bright-field image from the interference signals using known techniques, such as 4-phase or 5-phase modulation, or sinusoidal synchronous demodulation.

[0028] The device according to the invention also allows dark field imaging to be performed. In the configuration of the second synchronization mode, the illumination is synchronized with the rolling shutter so that the pixels exposed at each time step are not illuminated by the light backscattered by the object, and respectively reflected in the reference arm, but only by multi-scattered light. This configuration makes it possible to obtain absorption and / or phase contrast images.

[0029] The image processing module is configured to produce a dark field image from incoherent signals, i.e., from light multi-scattered by the object and not producing interference signals.

[0030] Importantly, the device according to the invention is configured to image an object or sample in both bright field and dark field, and to exploit the measurements obtained in both modes. The transition from one synchronization mode to the other is achieved by a simple adjustment of the time offset between the rolling shutter of the sensor and the scanning means, without any mechanical manipulation by an operator being necessary. Thus, from the same device that can operate in different bright field and dark field configurations, different types of full field images can be obtained of the same object or the same area of ​​the object.

[0031] Thanks to the device according to the invention, it is possible to easily switch from a full-field configuration, allowing measurements of the diffusion of structures with high intensity contrast, to a dark-field configuration, with offset detection, allowing phase contrast and / or absorption measurements. A sequence with different types of images can in particular be used to digitally correct movements of the imaged object or other disturbances.

[0032] Projecting a light line scanning the object to be imaged (such as the retina) at the same speed as the rolling shutter operates, with a constant time lag, during an acquisition, between the position of the shutter and the illumination line, allows high-resolution images to be obtained even in aberrant environments. Adjusting the rolling shutter lag and exposure time allows the angle of the phase gradient and the signal-to-noise ratio to be changed. When switching from the first to the second mode, the sensor exposure time is ideally increased. For example, in bright field, the lag is 0 and the exposure time can be 50μs. To switch to dark field, the lag can increase, for example, to 100μs, and the exposure time to 200μs.

[0033] The device according to the invention thus makes it possible, compared to state-of-the-art systems, to increase the imaging rate while increasing the detection sensitivity. A compromise between sensitivity and acquisition speed can be obtained by maintaining a degree of confocality in one transverse dimension while parallelizing the detection of pixels in the other transverse dimension with the light line.

[0034] The device according to the invention advantageously allows for easy implementation, as well as benefiting from multi-modality through dark field measurement.

[0035] In the device according to the invention, illumination and detection are decoupled, or independent, from each other. Thanks to the presence of a two-dimensional sensor with the rolling shutter, the detection line on the sensor is not in a fixed position, it is not necessary to "un-scan" the line under detection, that is to say to go back through the scanning system to cancel the introduced angle. It is then easy, with a simple time delay, to look outside the illuminated area of ​​the object to obtain the phase and absorption contrasts. This makes it possible to quickly switch from one synchronization mode to the other.

[0036] Thanks to the presence of a two-dimensional optical sensor (and not a one-dimensional sensor), the device according to the invention is robust with respect to the non-linearity of the scanning means, which may result from the mechanical constraints of these means, and to the optical aberrations introduced by them, which may respectively cause artifacts and a loss of spatial resolution. Due to the decoupling between illumination and detection, the non-linearities of the means for scanning the light beam can cause a time delay between illumination and detection and thus the loss of the signal. When there is no signal, it is then possible to directly observe the non-linearity of the scanning system, which is not possible in confocal or line imaging. Indeed, in confocal and line systems, since illumination and detection are coupled, the non-linearities of the scanning system cause a distortion artifact of the structure of the imaged object.

[0037] The time delay to switch from one mode to another is typically a few microseconds.

[0038] A high imaging rate is particularly important in the case of eye imaging, to overcome defects induced by eye movements.

[0039] High sensitivity allows the detection of less scattering structures, such as individual cells and their organelles.

[0040] The two-dimensional sensor can be, for example, a CMOS camera (complementary metal oxide semiconductor).

[0041] According to one embodiment, the synchronization module can be configured to operate alternately in the first synchronized mode and the second synchronized mode for each position of the light line in the field of view.

[0042] Alternatively, the synchronization module can be configured to operate in the first synchronized mode and the second synchronized mode for the entire field of view.

[0043] Advantageously, the image processing module may be configured to produce a phase contrast image from at least two dark field images.

[0044] Indeed, it is possible to obtain an enhanced phase contrast image by subtracting two images taken in dark field, also representing phase contrast images, with a time lag between the shutter and the positive or negative scanning means, respectively.

[0045] Alternatively or additionally, the image processing module may be configured to produce an absorption contrast image from at least two dark field images.

[0046] Indeed, it is possible to obtain an absorption contrast image by adding two images taken in dark field with a time lag between the shutter and the positive or negative scanning means, respectively.

[0047] According to an advantageous embodiment, the device according to the invention may further comprise an adaptive optics module configured to correct static and / or dynamic defects experienced by the wavefront of the illumination passing through the object.

[0048] With the adaptive optics module, static and / or dynamic defects imposed on the wavefront by the object or sample, e.g., the human eye, can be effectively corrected. Adaptive optics is particularly important when the object to be imaged is thick with large optical index variations, and / or has a non-planar surface. Correction must be performed quickly, in the case of dynamic aberrations, as for example in the case of the human eye in vivo. The correction allows maintaining a high resolution for the device even in the presence of defects or disturbances. Adaptive optics also ensures that the illumination remains line-shaped and does not spread, so as not to limit the contrast that can be achieved with the device by masking the phase signal by the direct signal of the spread line.It also ensures that the overlap between the size of the light line and the rolling shutter is optimal. The adaptive optics module thus guarantees both the correct dimension of the light line on the object to be imaged and the imaging quality of the structures of interest.

[0049] Advantageously, the illumination module may comprise a two-dimensional light source coupled to a line-generating lens.

[0050] Alternatively, the illumination module may comprise an extended light source coupled with a line-shaped spatial filter, or with a micromirror array (digital micromirror device, DMD).

[0051] The light source can be an infrared source.

[0052] Since infrared wavelengths are only very weakly absorbed by biological tissues, they are particularly suitable for imaging such tissues.

[0053] In one example, the light source may be a superluminescent diode (SLD). SLDs have the advantage of being relatively powerful (a few milliwatts) and having high directivity. In addition, their short temporal coherence length allows speckles, i.e., granular intensity distributions, to be attenuated.

[0054] In another example, the light source may be a light-emitting diode (LED), or a scanning laser.

[0055] The line-generating lens can be a cylindrical lens or a Powell lens.

[0056] A Powell lens ensures a homogeneous intensity distribution along the generated light line.

[0057] According to another aspect of the same invention, a method is provided for imaging a scattering object. The method is implemented by a device for interferometric imaging comprising a low coherence linear illumination module configured to generate a measurement beam and a linear section reference beam, interferometric means, a two-dimensional sensor comprising a rolling shutter, scanning means, a synchronization module and an image processing module, and in particular by an interferometric imaging device according to the present invention.

[0058] The method according to the invention comprises the following steps:

[0059] - illumination of the object with the measuring beam producing a line-shaped illumination in the field of view;

[0060] - generation of an interference signal resulting from interference between the backscattered measuring beam in the light line and the reference beam;

[0061] - scanning, by the scanning means, of the light line on the object in the field of view;

[0062] - synchronization, by the synchronization module, of the scanning means and the rolling shutter with a zero time shift in a first synchronized mode, in which the sensor is configured to capture the interference signal, and with a non-zero time shift determined in a second synchronized mode, in which the sensor is configured to capture a light signal, called incoherent, coming from a non-illuminated area of ​​the object;

[0063] - detection, by the sensor, of an interference signal and / or an incoherent signal;

[0064] - production, by the image processing module, of a bright field image from the interferometric signals and a dark field image from the incoherent signals.

[0065] Advantageously, the synchronizing and detecting steps may be performed alternately to produce a bright field image and to produce a dark field image successively.

[0066] It is indeed easy to switch from the first synchronized mode to the second synchronized mode by adjusting the time lag between the rolling shutter of the sensor and the scanning means. Bright field and dark field images can thus be obtained of the same object or the same part of the object, making it possible to combine the advantages of the two modes to obtain more information from the imaged object.

[0067] According to an advantageous embodiment, the method according to the invention may further comprise a step of producing a phase contrast image from at least two dark field images.

[0068] Alternatively or additionally, the method according to the invention may further comprise a step of producing an absorption contrast image from at least two dark field images.

[0069] The method according to the invention may also comprise a step of correcting static and / or dynamic defects experienced by the wavefront of the illumination passing through the object.

[0070] For this, an adaptive optics module can be implemented as described above.

[0071] The device and method according to the invention can be implemented in numerous applications requiring the imaging of scattering structures, objects or samples. The device and method according to the invention can be implemented in all fields where optical coherence tomography (OCT) is used and useful, namely, medical diagnosis of surgical samples, ophthalmological medical diagnosis (for example, of the cornea or retina) in vivo, and imaging of biological samples. This last application is particularly important in pharmacology, in particular for the imaging of organoids.

[0072] The device and method according to the invention can in particular be implemented in full-field OCT (FF-OCT), in the time or spectral domain, or in source-scanning Fourier OCT (FD-OCT).

[0073] The device and method according to the invention can also be used for imaging various materials.

[0074] More particularly for retinal imaging, the device and method according to the invention make it possible to observe most of the structures of interest in a retina, whether they are highly backscattering, such as cones and rods, or weakly scattering in bright field, or less scattering in dark field, such as ganglion cells and blood capillaries. The high achievable frame rates and the wide imaged field of view make it possible to meet a medical need to scan the largest possible area while limiting the imaging time for the patient and thus reducing movement artifacts. The high frame rate also makes it possible to image blood flow in capillaries present in the retina. Description of figures and embodiments

[0075] Other advantages and characteristics will appear on examining the detailed description of non-limiting examples, and the appended drawings in which:is a schematic representation of a non-limiting exemplary embodiment of an interferometric imaging device according to the invention;is a schematic representation of an example of an illumination module of the device according to the invention;shows a schematic representation of a first synchronization mode implemented by the device according to the invention; andshows a schematic representation of a second synchronization mode implemented by the device according to the invention.

[0076] It is understood that the embodiments described below are in no way limiting. In particular, all the variants and embodiments described can be combined with each other if there is no technical obstacle to this combination.

[0077] In the figures, elements common to several figures may retain the same reference.

[0078] This is a schematic representation of a non-limiting embodiment of an interferometric imaging device according to the present invention.

[0079] The device 1, as shown in the, can be implemented to carry out the steps of an interferometric imaging method according to the present invention. Exemplary embodiments of the device and the method according to the invention will be described subsequently.

[0080] The device 1 comprises a linear illumination module 2, interferometric means, scanning means 12 and a two-dimensional image sensor 14.

[0081] The linear illumination module 2 comprises a low-coherence light source. The module 2 makes it possible to produce a light beam of linear cross-section, and thus a line-shaped illumination in the plane of the object 20 or the sample to be imaged. An example of such an illumination module is illustrated in the.

[0082] The illumination module 2, as shown in the, comprises a fibered superluminescent diode (SLD) 3, a collimator 4 and a line-generating lens 5. The line-generating lens 5 is preferably a Powell lens. The lens 5 may also be a cylindrical lens.

[0083] The SLD 3 is a two-dimensional source. It has broadband emission and a short temporal coherence length. For example, it can emit at a wavelength of 850 nm with a spectral bandwidth of 55 nm.

[0084] Alternatively, the output of the light source can also be unfibered. In this case, if the outgoing light beam is already parallel, a collimator is not necessary.

[0085] Beam 7 from SLD 3 passes through collimator 4 and Powell lens 5, which converts the collimated beam into a diverging beam with a linear cross-section. The converted beam has an opening angle in one direction but remains collimated in the orthogonal direction.

[0086] Illustrates the shape of the light beam 7 coming out of the illumination module 2 in the two planes (x, z) and (y, z), as well as its linear section 25 in the plane (x, y).

[0087] Still with reference to the, an achromatic doublet 6 positioned at the output of the illumination module 2 then makes it possible to limit the illumination field in the direction in which the beam is divergent, in order to minimize power losses. By superimposing the focal planes of the cylindrical lens 5 and the doublet 6, a collimated light line is formed after the doublet 6.

[0088] In order to ensure homogeneous illumination on the edges of the light line, and as illustrated in the, a slit 13 can be placed in the image focal plane of the achromatic doublet 6, this focal plane being conjugated to the plane of the object or sample to be imaged.

[0089] The device 1 also comprises interferometric means. In the embodiment shown in the, the interferometric means comprise a beam splitter cube 8 and a movable mirror 9. The line-shaped light beam 7 from the illumination module 2 is split by the splitter cube 8 into a measuring beam 10 and a reference beam 11, forming a measuring arm and a reference arm, respectively, of the interferometer. The measuring beam 10 is directed towards the object 20 and focused thereon using a lens 21 and an objective 23, so that the object is illuminated with the light line generated by the illumination module 2. The light scattered by the object 20 passes back through the splitter cube 8.

[0090] The movable mirror 9 allows the length of the reference arm to be varied. The reference beam 11, focused on the mirror 9 by the lens 21 and an objective 24, is reflected by the mirror 9 towards the splitter cube 8. The reference beam 11 then interferes with the light diffused by the object 20.

[0091] In the embodiment shown in the, the interferometric means are arranged according to the so-called Linnik configuration.

[0092] Of course, other configurations for the interferometer are possible.

[0093] The light line is scanned over the object 20 to be imaged, in the field of view, using scanning means. For this, the light beam 7 emitted by the illumination module is directed towards a galvanometric mirror 12. This allows the measurement beam 10 to be directed towards the object 20 at a variable angle, allowing the entire field of view to be scanned with the light line formed by the illumination module 2. The reference beam is also scanned by the galvanometric mirror 12 on the reference mirror 9.

[0094] The scanning amplitude can be adjusted to provide a specific field of view. The scanning frequency of the galvanometer mirror is the same as the frequency for acquiring full-field images over the field of view. For example, to acquire images over the entire field of view, the scanning speed might be 100 Hz, and for images over half the field of view, the scanning speed might be 200 Hz.

[0095] The illumination of the object 20 must be line-shaped in the image focal plane, and the galvanometric mirror 12 must be in the pupil plane so that the angle of the mirror 12 allows a change in position of only the light line on the object.

[0096] Two lenses 17, 18 arranged on either side of the galvanometric mirror 12 make it possible to move from an image plane to a pupil plane.

[0097] The device 1 according to the invention comprises a two-dimensional sensor 14 configured to detect interference signals as well as light scattered by the object. The light signals are focused onto the detector using a lens 22. The sensor 14 comprises a rolling shutter. This shutter is an electronically controllable shutter. In particular, it is possible to adjust the width of the shutter in order to define the number of rows or lines of pixels of the sensor 14 that are exposed. The exposure time per row of pixels of the sensor 14 is constant. The speed of the rolling shutter is defined by the frame rate of the sensor 14.

[0098] Sensor 14 is for example a CMOS camera equipped with a rolling shutter.

[0099] The device 1 according to the invention comprises a synchronization module (not shown) for managing the synchronization mode between the scanning means and the rolling shutter of the sensor 14.

[0100] A first and second synchronization mode are schematically illustrated in Figures 3 and 4, respectively. In Figures 3 and 4, the field of view of the sensor is represented by a 15 x 9 pixel matrix 31. The resulting light line from the superposition of the backscattered light 32 and the light reflected by the reference arm on the sensor is illustrated by a gray background. The rolling shutter exposing the three rows of pixels is illustrated by a rectangle 33 with a thick line.

[0101] In a first synchronization mode, the time lag between the scanning means and the rolling shutter is zero. In this case, the sensor captures the area of ​​the object illuminated by the light line, it then detects the interference signal between the light scattered by the object and the reference beam, for each illuminated line-shaped area on the object. This is bright-field detection.

[0102] The operating principle of the first synchronization mode is illustrated in the.

[0103] In the first synchronization mode, the light directly backscattered, in line form, by the object as well as the reference beam are detected by the rows of pixels 31 of the sensor exposed by the rolling shutter. To synchronize the rolling shutter with the scanning means, and in particular the galvanometric mirror, a TTL (transistor-transistor logic) type electronic trigger signal can be generated and sent to the sensor, synchronously with a linear voltage signal, of sawtooth type generated and sent to the galvanometric mirror, for example with an analog output card.

[0104] To filter out multi-scattered, unfocused light, the center of the backscattered light line must coincide with the center of the pixel rows exposed by the shutter.

[0105] In the example shown in the, the backscattered light line 32 illuminates three rows of pixels 31 on the sensor. The rolling shutter speed is determined i per row. The exposure of the first row of pixels begins at time t0, which corresponds, for example, to the rise of the trigger signal. The exposure of the next row of pixels begins after t i , and so on until the last row of pixels in the image.

[0106] Exposure time exp is, for example, 50 µs, for an acquisition frequency of 200 Hz. This corresponds to leaving each line exposed for the equivalent of 10 rows of pixels. For an illumination line width of 10 rows of pixels on the 14 sensor, choosing an exposure time of 50 µs is optimal. For a field of view of 1000 rows of pixels, this corresponds to an acquisition time of 5.05 ms per image.

[0107] A complete image is acquired when all the pixels corresponding to the field of view, i.e. the 1000 rows, have been exposed, i.e. in 5.05 ms. To obtain an image from the interference signals, known methods (such as FD-OCT or TD-OCT) can be implemented.

[0108] The galvanometric mirror returns to its initial position before acquiring a new image.

[0109] The images obtained using this first synchronization mode are reflectance contrast images.

[0110] In a second synchronization mode, the time lag between the scanning means and the rolling shutter is non-zero. In this case, the sensor captures an incoherent light signal, coming from an unilluminated area of ​​the object. This is the dark field detection of the light multi-scattered by the object.

[0111] The operating principle of the second synchronization mode is illustrated in the.

[0112] In the second synchronization mode, the light directly backscattered, in a line shape, by the object as well as the reference beam are not detected by the sensor. Therefore, the sensor does not detect interference signals.

[0113] Here, the rolling shutter of the sensor is temporally offset relative to the galvanometric mirror (and thus the light line 32 on the sensor) by a time of 3t i . In the example shown in the, the rolling shutter 33 is ahead of the illumination. In the plane of the figure, the exposed pixels are then below the illuminated pixels.

[0114] The backscattered light line 34 illuminates two rows 32 of pixels 31 on the sensor, and the rolling shutter speed is determined iper row. The exposure of the first row of pixels begins at time t0, which corresponds, for example, to the rise of the trigger signal. The exposure of the next row of pixels begins after t1, and so on until the last row of pixels in the image. The rolling shutter also exposes two rows 35 of pixels, offset by three rows relative to the illuminated rows 32.

[0115] Of course, the rolling shutter can also be delayed relative to illumination.

[0116] The images obtained using this second synchronization mode are dark field images, from which phase contrast and absorption images can be generated.

[0117] In the device according to the invention, the synchronization module is configured to operate in the first and second synchronization modes, combining bright field and dark field imaging to improve the quality of imaging by the device.

[0118] To obtain an enhanced phase contrast image, two dark-field images, also representing phase contrast images, can be subtracted. One of the two original images is captured with a positive time shift between the shutter and the scanning means, and the other original image with a negative time shift. The time shifts preferably have the same value, but only change sign.

[0119] To obtain an absorption contrast image, two dark-field images can be added together. One of the two original images is captured with a positive time lag between the shutter and the scanning means, and the other original image with a negative time lag.

[0120] Scanning and synchronization parameters, such as the electronic trigger signal, the width and speed of the rolling shutter, or the various time shifts implemented, can be controlled via a user interface.

[0121] Advantageously, the device according to the invention may also comprise an adaptive optics module. The adaptive optics module makes it possible to measure and correct an aberrated wavefront, for example due to movements of the imaged object, and therefore to improve the resolution of the images.

[0122] The adaptive optics module comprises a light source (called an analysis source) such as a fibered superluminescent diode, a wavefront sensor, for example of the Shack-Hartmann type, and a deformable mirror. The wavelength or spectral band of the analysis source is preferably different from that of the source used for imaging; otherwise, spectral filters can be used to separate the sources. The beam from the light source is directed towards the object. The deformed wavefront of the light backscattered by the object is directed towards the deformable mirror and reflected towards the wavefront sensor which measures the wavefront deformation. The deformable mirror, which functions as a wavefront corrector, is controlled in real time to correct the wavefront detected by the wavefront sensor.

[0123] In this case, the adaptive optics module must be arranged in the measuring arm of the interferometer.

[0124] The wavefront is corrected for both the measurement beam going to the object and the scattered beam from the object, with the wavefront corrector placed in a path common to both illumination and imaging.

[0125] The device according to the invention also comprises an image processing module.

[0126] The image processing module produces a bright-field interference image from the interference signals. Different methods can be used to obtain an interference image from an interference signal, by taking successive images at different positions of the reference mirror, such as four-phase imaging, five-phase imaging, or synchronous detection. These methods are known and will not be discussed in further detail.

[0127] The image processing module also allows producing a dark field image from the incoherent signals.

[0128] The image processing module comprises at least one computer, a central or computing unit, a microprocessor, and / or suitable software means.

[0129] An exemplary embodiment of a method for imaging a scattering object will be described below. The method may be implemented, for example, by the imaging device according to the example described in relation to Figures 1 and 2.

[0130] The method according to this embodiment comprises the following steps: illumination, by the linear illumination module, of the object with a light line in the field of view; generation of an interference signal resulting from interference between the backscattered measuring beam in the light line and the reference beam; scanning, by the scanning means, of the light line on the object in the field of view; synchronization, by the synchronization module, of the scanning means and the rolling shutter with a zero time shift in the first synchronized mode, in order to detect, with the sensor, an interference signal and to acquire, with the image processing module, a bright field image.

[0131] The acquired brightfield image(s) can then be optimized, in terms of image quality, to find the correct time offset between the line scan and the rolling shutter. This offset can be different from one sample to another.

[0132] When the object to be imaged is the retina, this step is important to ensure that the imaging parameters are correct, and in particular that the photoreceptors can be imaged by positioning themselves in the correct plane. This also allows the optimal time shift to be set for dark-field imaging.

[0133] In particular, the optimization step makes it possible, in bright field mode, to ensure that the rolling shutter is centered on the light line at all times and that the width of the rolling shutter is identical to that of the line. This condition is verified when the signal on the bright field image obtained is homogeneous over the entire image and maximum. The time shift is zero in this case.

[0134] Optimal adjustment of the brightfield imaging mode allows the introduction of a controlled time shift for phase contrast imaging obtained from the darkfield imaging mode.

[0135] Thus, if the bright field mode is set correctly, the same time shift value applied in advance or delay effectively results in a rolling shutter shifted forward (advance) or backward (delay) with the same spatial shift at each instant, symmetrically about the center of the bright line. Without this symmetry, the symmetry in dark field mode between the image obtained with an advance of the rolling shutter and the image obtained with a delay of the rolling shutter would be compromised.

[0136] This optimization or adjustment step can be carried out during a device calibration phase.

[0137] Optionally, in order to correct static and / or dynamic defects experienced by the illumination wavefront passing through the object or sample and ensure a well-defined light line on the object as well as high imaging resolution, an adaptive optics module can be implemented as described above.

[0138] In a subsequent step, a positive time shift can be applied to the scan, by synchronizing, with the synchronization module, the scanning means and the rolling shutter with a positive time shift in the second synchronized mode, in order to detect, with the sensor, an incoherent signal and produce, with the image processing module, a dark field image.

[0139] The acquired dark field image(s) can be optimized, in terms of image quality, to find the correct time shift.

[0140] Indeed, in dark field mode, the offset required to obtain optimal contrast depends on the structures to be observed. It is possible to carry out a parametric study by acquiring image sequences of the same region for the same exposure time, by varying the time offset between the illumination line and the rolling shutter, in order to be able to observe the evolution of the contrast and sharpness of the structure to be imaged as a function of the time offset. From this data, the optimal time offset can be extracted to maximize the contrast of the structure of interest.

[0141] The same steps can be implemented to acquire one or more dark-field images with a negative time shift of the same value as the positive time shift.

[0142] Between steps to acquire darkfield images or brightfield images, it may be useful to return to brightfield mode or darkfield mode, respectively.

[0143] For example, during acquisitions, the eye, for example, fixes a target and is subject to involuntary movements inducing relative shifts between the acquired images. The images must therefore be registered in post-processing. It may then be easier to estimate the relative shifts between the images from the bright-field images than from the dark-field images. The estimation of the shift of each bright-field image can also be used to extrapolate the shift of the successive dark-field image to that just estimated with the bright-field image.

[0144] Another example is that for some retinal structures, such as photoreceptors, it may be useful to acquire images with absorption contrast and phase contrast. It is then possible to chain together bright-field image acquisitions (for example, a hundred or so) and then dark-field image acquisitions. Alternatively, image acquisitions can be interspersed between bright-field and dark-field. The advantage of this alternative is that the same structure is imaged at very close intervals, potentially allowing the study of retinal cell dynamics.

[0145] According to one embodiment, the method further comprises a step of producing a phase contrast image from at least two dark field images. To do this, an image acquired with a positive time shift and an image acquired with a negative time shift are subtracted from each other.

[0146] As an example, Figure 6 shows a phase contrast image obtained with a method according to the embodiment described above. In the image, blood vessels, capillaries and fibers can be distinguished.

[0147] When alternating between the dark field modes corresponding to the positive and negative time shifts, it is then possible to consecutively subtract the images from the acquisition two by two to obtain a so-called "split detection" image sequence with maximized phase contrast. If the initial acquisition sequence with the alternation of the dark field modes with the positive and negative time shifts is, for example, 800 images / s, the effective rate of the "split detection" image sequence, obtained from two successive images of the initial sequence, is half as low, i.e. 400 images / s.

[0148] When the method according to the invention is implemented for retinal imaging, it is appropriate to check, when switching on the light source of the illumination module, and possibly the source of the adaptive optics module, that the sources comply with ocular safety.

[0149] Of course, the invention is not limited to the examples which have just been described and numerous adjustments can be made to these examples without departing from the scope of the invention.

Claims

Device (1) for interferometric imaging of a scattering object (20), comprising:a low coherence linear illumination module (2) configured to generate a measurement beam (10) and a reference beam (11) of linear section, the measurement beam (10) producing a light line-shaped illumination of the object (20) in a field of view;interferometric means (8, 9) configured to generate an interference signal resulting from interference between the measurement beam backscattered in the light line and the reference beam;a two-dimensional sensor (14) comprising a rolling shutter;scanning means (12) configured to move the light line over the object (20) in the field of view;a synchronization module configured to operate in a first synchronized mode with a zero time offset between the rolling shutter and the scanning means, in which the sensor (14) is configured to capture the interference signal, and in a second synchronized mode with a determined non-zero time offset between the rolling shutter and the scanning means, in which the sensor (14) is configured to capture a light signal, called incoherent, coming from a non-illuminated area of ​​the object (20); andan image processing module configured to produce a bright field image from the interference signals and to produce a dark field image from the incoherent signals.; Device (1) according to the preceding claim, characterized in that the synchronization module is configured to operate alternately in the first synchronized mode and the second synchronized mode for each position of the light line in the field of view. Device (1) according to claim 1 or 2, characterized in that the image processing module is configured to produce a phase contrast image from at least two dark field images. Device (1) according to any one of the preceding claims, characterized in that the image processing module is configured to produce an absorption contrast image from at least two dark field images. Device (1) according to any one of the preceding claims, characterized in that it further comprises an adaptive optics module configured to correct static and / or dynamic defects experienced by the wavefront of the illumination passing through the object (20). Device (1) according to any one of the preceding claims, characterized in that the illumination module (2) comprises an infrared light source (3). Device (1) according to any one of the preceding claims, characterized in that the illumination module (2) comprises a two-dimensional light source (3) coupled to a line-generating lens (5). Device (1) according to any one of the preceding claims, characterized in that the sensor (14) is a complementary metal oxide semiconductor sensor. A method for imaging a scattering object (20), the method being implemented by a device (1) for interferometric imaging comprising a low coherence linear illumination module (2) configured to generate a measuring beam (10) and a reference beam (11) of linear section, interferometric means (8, 9), a two-dimensional sensor (14) comprising a rolling shutter, scanning means, a synchronization module and an image processing module, the method comprising the following steps:illuminating the object with the measuring beam (10) producing a light line-shaped illumination in the field of view;generating an interference signal resulting from interference between the backscattered measuring beam in the light line and the reference beam;scanning, by the scanning means, the light line over the object (20) in the field of view;synchronization, by the synchronization module, of the scanning means and of the rolling shutter with a zero time shift in a first synchronized mode, in which the sensor (14) is configured to capture the interference signal, and with a non-zero time shift determined in a second synchronized mode, in which the sensor (14) is configured to capture a light signal, called incoherent, coming from a non-illuminated zone of the object (20);detection, by the sensor (14), of an interference signal and / or an incoherent signal;production, by the image processing module, of a bright field image from the interferometric signals and a dark field image from the incoherent signals.; Method according to claim 9, characterized in that the synchronization and detection steps are carried out alternately to produce a bright field image and to produce a dark field image successively. Method according to claim 9 or 10, characterized in that it further comprises a step of producing a phase contrast image from at least two dark field images. Method according to any one of claims 9 to 11, characterized in that it further comprises a step of producing an absorption contrast image from at least two dark field images. Method according to any one of claims 9 to 12, characterized in that it further comprises a step of correcting static and / or dynamic defects suffered by the wavefront of the illumination passing through the object.

Images