DE102010044826B4 - Detector and measuring device and method for determining the thickness of a sample - Google Patents

Abstract

Detector (8) to which backscattered light from a sample (4) positioned in an OCT arrangement (2) is fed, for the simultaneous, two-dimensional spatially resolved registration of a plurality of spectrally resolved intensity distributions (25) of the backscattered light, each intensity distribution ( 25) is associated with a discrete point on the sample surface (10), the detector (8) having imaging optics (13) and a two-dimensional light sensor (14) onto which the intensity distributions (25) are imaged, characterized in that the imaging optics (13) comprises a microlens array (15) and a dispersive optical element in the form of a volume grating (16), wherein the volume grating (16) is a diffractive optical element in the form of a computer generated hologram.

Description

The invention relates to a detector to which backscattered light is supplied from a sample positioned in an OCT array. Moreover, the invention relates to a measuring device for determining the thickness of a sample, wherein the measuring device has an OCT arrangement and such a detector. Finally, the invention also relates to a method for determining the thickness of a sample.

Optical coherence tomography (OCT) is a method of measuring distances based on the interference of short coherent light. It is also used in the medical field, there z. As for measuring the eye length in ophthalmology, but also often as an imaging method for high-resolution, three-dimensional imaging of highly diffuse-scattering objects.

The principle of OCT consists of splitting a broadband, short-coherent light beam into a measuring beam and a reference beam. Following the principle of a Michelson interferometer, the two partial beams are combined again. In the classical implementation of Time Domain OCT (TD-OCT), the movement of the reference mirror or the detector, which is usually designed as a highly sensitive photodiode, extracts the depth information contained in the interferogram along the direction of the measuring beam. Such a measurement is also called an A-scan.

Since the depth information is also contained in the spectrum of the interferogram, it can also be reconstructed by measuring just that spectrum. This so-called Fourier Domain OCT (FD-OCT), which is also called Frequency Domain OCT or "Spectral Radar", allows the recording of many (up to several thousand) using a sensitive grating spectrometer and a line scan camera as a detector ) A-scans per second.

A B-scan is the combination of several linearly successively performed A-scans in order to obtain a section through the considered surface. The so-called "en-face" OCT accomplishes this by using ordinary rectangular CCD chips as detectors instead of the line scan cameras. The image of the spectrum of the measuring beam, which is not designed here as a bundle of light, but as a fan, allows the simultaneous recording of a complete B-scan. This allows you to take several (typically 20 to 60) B scans per second.

A spatially resolved measurement (C-scan) usually requires scanning the surface in two dimensions. This is usually accomplished by deflecting the measuring beam (and analogously the returning signal) by means of two xy scanning mirrors. This technique, similar to the excellent depth resolution of the single A-scans in the range of a few microns, allows to obtain a similar lateral resolution, so that real three-dimensional data sets with isotropic resolution can be created. The disadvantage here is that the scanning of the surface for mechanical and electrical reasons, regardless of the technology requires a certain amount of time. In the case of fast-moving DUTs, large measurement uncertainties (artefacts) are thereby generated due to the special properties of the FD-OCT reconstruction algorithm. However, when investigating living biological objects, a necessary "stand-still" is not readily possible. The necessary for the deflection of the beam scanner also require a complex electronic control, which make the overall system more expensive and therefore precludes a proliferation of such systems.

A special application of OCT is the laterally resolved measurement of the thickness of a sample of highly scattering material, in particular biological tissue. The sample may be, for example, the cornea of the eye, ie the outermost layer of the eye, which decisively contributes to the total refractive power of the eye. The knowledge of the spatially resolved thickness of the cornea is of essential importance for the planning of refractive procedures such. For example, the LASIK, but also for the early diagnosis of degenerative processes that lead to a change in the thickness of the cornea (dystrophy, keratoconus).

Due to the fast involuntary eye movements (saccades, up to 200 Hz) on the one hand, and the long measurement time for C-scans on the other hand, as well as the high price of devices with correspondingly fast electronics, which allow the occurrence of artifacts within acceptable limits Devices for full three-dimensional measurement of the anterior segment of the eye are not widely available on the market.

Fourier domain OCT methods are for example in US2009 / 0015842A1 and DE102004037479A1 described. On an interference principle based detectors are also off DE 4108944C2 and DE4309056A1 known. All methods and detectors specified there are comparatively expensive in terms of design or it requires a great deal of effort to obtain suitable measurement results.

An object of the present invention is to obtain the thickness of a sample in a particularly simple manner.

This object is achieved by a detector according to claim 1 or by a measuring device according to claim 3 or by a method according to claim 4.

The advantages and embodiments explained below in connection with the method also apply mutatis mutandis to the devices according to the invention and vice versa.

The detector to which backscattered light is supplied by a sample positioned in an OCT arrangement is designed according to the invention for simultaneous, two-dimensionally spatially resolved registration of a plurality of spectrally resolved intensity distributions of the backscattered light, each intensity distribution being associated with a discrete point on the sample surface.

The measuring device is designed to determine the thickness of a sample at a plurality of discrete points with a single receptacle, and for this purpose comprises an OCT arrangement in which the sample is positioned, a detector, as indicated above, and an evaluation unit, designed for detection the depth information for these points from the spectral data thus obtained.

The corresponding method for determining the thickness of a sample at a plurality of discrete points with a single image is characterized in that the sample is positioned in an OCT arrangement, that backscattered light from the sample is fed to a detector, the spectrally resolved intensity distributions of the backscattered light associated with a plurality of discrete points on the sample surface registered simultaneously and two-dimensionally spatially resolved, and that from the spectral data thus obtained depth information for these points are determined.

Especially advantageous is the use of the above-mentioned detector or the measuring device or the method for measuring the thickness of the cornea of the human eye. Other applications are generally the determination of the thickness of optically more or less transparent layers, for. B. of paint or other cover layers, plastics, etc.

The invention is based on the finding that in the measurement of the thickness of the cornea in most cases no isotropic resolution is necessary in order to obtain useful diagnostic information. It is sufficient to know the thickness of the cornea (pachymetry) at a few, but sufficiently many (eg a few hundred) discrete points, in particular for surgical planning and for the diagnosis of degradation processes due to systemic underlying diseases. If one sets these measured thicknesses in relation to the height data of the cornea (topography) determined by other methods, for example placido-keratometry, a sufficiently accurate model for describing the cornea is obtained.

An essential idea of the invention is to provide a technique by means of which information about the thickness of a sample can be obtained in a particularly simple manner. Instead of trying to obtain depth information in the course of an imaging OCT method, a pure measurement method is proposed. In other words, the goal is not a particularly good imaging in the sense of an isotropic imaging of the sample. Rather, a simultaneous measurement is made of the thickness of the sample, at several discrete points of the sample surface.

For this purpose, it is provided that the light backscattered by a sample positioned in an OCT arrangement is supplied to a detector provided specifically for this measuring method. The detector assumes the function of a plurality of simultaneously and in parallel operating spectrometer. It records the spectrally resolved intensity distributions of the backscattered light originating from several non-adjacent points on the sample surface and performs a simultaneous, two-dimensional spatially resolved registration of these spectra. This allows each registered spectrum to be assigned to a specific point on the sample surface. From the registered spectra, the depth information for these points is determined with the aid of an evaluation unit. Thus, the thickness of the sample can be determined at several selected points with a single shot (exposure). The thickness determination is in other words simultaneously, d. H. at the same time for all measuring points. From the depth information, a depth profile of the sample can then be created. The measurement results can also be advantageously combined with results of other measurement methods.

A "discrete measuring point" or "discrete point" is understood to be a finely extended area on the sample surface. The size and arrangement of the area is determined by the respective subaperture. In the method according to the invention, in other words, each registered spectrally resolved intensity distribution is assigned to a specific area of the sample surface. These finite areas can be considered as discrete points of measurement. As a result of the device according to the invention, a measured value of the thickness is assigned to each of the discrete points defined by respectively suitable coordinates. This corresponds to an average value of the actual thickness over the considered area. For reasonably smooth ("smooth") surfaces this approximation is allowed.

Advantageous embodiments of the invention are specified in the subclaims.

According to the invention, the detector comprises an imaging optics for discrete spectral decomposition and a two-dimensional light sensor, on which the intensity distributions are imaged. The structural design is therefore comparatively simple. In this case, a low-cost CCD sensor in the form of a two-dimensional CCD array can be used as the light sensor. Expensive sensors are not needed. It is particularly advantageous if only a single two-dimensional light sensor is used, in other words all spectra are registered on a single sensor. Such an embodiment can be realized structurally particularly simple and inexpensive.

The imaging optics according to the invention comprises a microlens array and a dispersive optical element. The dispersive optical element serves to realize the spectral decomposition of the (white) light. The microlens array serves to focus the backscattered light supplied to the detector onto the dispersive optical element. The dispersive optical element is designed as a micro-optical volume grating. Thus, a transmissive arrangement can be realized in a comparatively simple manner. Instead of a diffraction grating, a plurality of microprisms could also be used in such a case. As an alternative to a transmitted-light arrangement, it is possible to provide an arrangement which allows reflection to function, for example using an inclined, reflecting grating.

According to the invention, a diffractive optical element (DOE), in particular in the form of a computer-generated hologram (CGH), is used as a volume grating, so that a sufficient diffraction efficiency also results in the transmissive arrangement.

In a particularly preferred embodiment of the invention, the imaging optics has a diaphragm for masking parts of the intensity spectrum. The hidden parts are preferably the undiffracted portions of the light (zeroth order) and one side of the symmetrical spectrum, respectively. In other words, for each spectrum of a point on the sample surface through the aperture, one-half of the spectrum symmetrically diffracted by the volume lattice is blanked out. This is particularly advantageous because the desired depth information is completely contained in the remaining part of the spectrum and can be exploited very effectively by hiding the available space on the light sensor.

In summary, a true three-dimensional detector is provided with a two-dimensional light sensor. In this case, the xy spatial resolution of the light sensor is not used in the first place, but rather the possibility of registering spectral data containing depth information on the light sensor, spatially resolving and thus associated with the individual discrete measurement points on the sample surface.

With the aid of the invention, a true three-dimensional detector is thus provided in a comparatively simple manner, which makes it possible to simultaneously record (simultaneously) all three dimensions of the object to be examined, in this case the sample, with a single image. For this purpose, a simultaneous measurement takes place at individual, spaced-apart points of the sample.

• 1 eine Übersichtsdarstellung der Meßanordnung,
• 2 eine Detaildarstellung eines ersten Detektors,
• 3 eine Detaildarstellung eines zweiten Detektors,
• 4 eine Darstellung der erhaltenen Aufnahme auf dem Lichtsensor.

Embodiments of the invention are explained below with reference to the drawings. Hereby show:

• 1 an overview of the measuring arrangement,
• 2 a detailed representation of a first detector,
• 3 a detailed representation of a second detector,
• 4 a representation of the image obtained on the light sensor.

All figures show the invention only schematically and with its essential components. The same reference numerals correspond to elements of the same or comparable function.

In the embodiment described below, based on the principle of FD-OCT, the depth information in the spectrum of an interferogram is used to obtain a complete measurement of the thickness of the cornea of an eye at discrete points with a single shot.

As in 1 shown, comprises the measuring device according to the invention 1 a FD-OCT arrangement 2 in the form of a fiber-coupled Michelson interferometer. It finds a spectrally broadband, short-coherent light source 3 Use, for example, a superluminescent diode (SLD). As a sample 4 serves optically scattering material, here biological tissue in the form of the cornea of an eye. The sample defines the first interferometer arm (measuring arm) 5 , The second interferometer arm ( reference arm) 6 is powered by a movable reference mirror 7 Are defined.

According to the basic principles of optical coherence tomography, only those portions of light returning from the sample 4 interfere with the reference light having traveled an optical path length within the coherence length of the light portions in the reference arm 6 lie. The interferometer thus provides at its output a specific interference signal. If this is spectrally resolved, the depth information can be calculated from it. According to the invention, the interference signal does not become a normal spectrometer but a detector 8th supplied, the operation of which will be explained in more detail below.

Since the sample is the cornea of an eye, is in the measuring arm 5 a diffractive optical element (DOE) 11 is inserted, which defines the plane wavefront, which is defined by the collimating optics 12 from the light source 3 generated and by a beam splitter 9 in the measuring arm 5 and the reference arm 6 is divided, transformed into a spherical wave. Assuming that the cornea 4 essentially has a spherical geometry, it can thus be achieved when positioning the eye at the point in which the radius of the wavefront corresponds to the mean corneal radius r _CORNEA that all portions of the measuring beam almost perpendicular to the surface 10 the cornea 4 to meet. Such an arrangement is in DE 103 42 175 A1 described. There, however, only the wavefront was evaluated. In the present invention, however, the returning signal passing through the beam splitter 9 recombined and the detector 8th is supplied to measure the vertical thickness of the cornea 4 used at discrete points.

The detector 8th is designed as a two-dimensional grating spectrometer and shows how out 2 can be seen, an imaging optics 13 and a light sensor 14 on. The imaging optics 13 comprises in transmission arrangement a microlens array 15 , a volume grid 16 and an aperture 17.

The microlens array 15 includes a first microlens array 18 , Here are the microlenses 19 of the first microlens array 18 arranged in a rectangular grid of n × m (eg 30 × 30) microlenses and positioned adjacent to each other so that they achieve the highest possible area coverage. The microlenses 19 have a focal length of a few millimeters. In addition, a second, in the beam direction 21 spaced microlens array 22 intended. This is in the arrangement of a Kepler telescope behind the first microlens array 18 , The two microlens arrays 18 . 22 In other words, they are positioned relative to each other such that those amounts of light that are incident on the subapertures 24 fall, each one of the microlenses 19 of the array 18 belong, bundled and on the second microlens array 22 be imaged. The second microlens array 22 serves the beam shaping and collimates the beam 23 then on the volume grille behind it 16 , Instead of the second microlens array 22 In a particularly simple embodiment, one can also be in the focus of the first microlens array 18 arranged slot mask can be used (not shown).

The volume grid, designed as Computer Generated Hologram (CGH) 16 , which splits the light into its frequency components due to the diffraction at the grating, throws it onto a two-dimensional CCD chip 14 , which serves as a light sensor. Previously, however, with the aid of an aperture arranged in the beam path 17 the zeroth diffraction order and the conjugate portions of the respective neighboring subaperture hidden. This is used, among other things, for the particularly effective utilization of the available chip area.

As a result, every subaperture becomes 24 a spectrum 25 on the surface 26 of the CCD chip 14 projected. In 2 is such an intensity distribution I = f (λ) shown symbolically. Because the origin of the beam from a certain point of the sample surface 10 is known, this can be the thickness of the sample 4 be determined at this point.

The resolution of the detector 8th more precisely the number of a single subaperture 24 associated pixels, is through the spectral width of the light source 3 and determined by the scanning of the spectra. Typical values for the number of subapertures 24 According to the invention are between 3 and 30 in each direction, so that with the inventive arrangement recorded at discrete areas of the cornea depth-resolved spectra for 9 to 900 points can be detected simultaneously. To increase the light output in each case several lines of the CCD chip 14 for the projection of the spectrum 25 a subaperture 24 used.

The arrangement of the spectra 25 on the CCD chip 14 is preferably such that the available area of the CCD chip 14 is essentially fully utilized. In addition to a maximum area filling at the same time a minimum crosstalk and the largest possible number of pixels per subaperture 24 be given. This includes non-orthogonal arrangements of the spectra on the CCD chip 14 one.

A simple arrangement of the spectra 25 on the CCD chip 14 shows 4 , For clarity, there are only four subapertures 24 represented, wherein in the two lower the projection 27 the spectra 25 on the CCD chip 14 are indicated.

The achievable measurement depth is with the number of pixels for the spectrum 25 a single subaperture 24 connected. An increase in the number of subapertures 24 while maintaining the size of the CCD chip 14 results in a reduction of the measurement depth according to the Nyquist theorem while maintaining the resolution. The usable depth of measurement over the entire depth range of the cornea 4 can be extended by the distance of the reference mirror 7 adapted to the geometry of the measurement object (eg dynamic method), as is required anyway for the compensation of the different corneal _radii r _CORNEA .

With a target resolution of 10 μm, Z _max = (ΔL / 2) × (N / 2) with a number of pixels per subaperture 24 of 200, a maximum measuring range of 500 .mu.m, where ΔL is the resolution and Z _{max is} the maximum axial measuring range (measuring depth). This represents the minimum to the thickness of the cornea 4 in one single measurement. The decisive factor is the presence of CCD chips 14 with enough pixels. Conventional CCD chips may be used to practice the invention 14 to be used with a pixel count of 1600 to 4000 in the direction of greater expansion. CCD chips 14 with larger numbers of pixels allow an enlargement of the measuring window while maintaining the resolution, or correspondingly an increase in resolution at the same measuring window.

When using a volume grating 16 may be a particularly simple transmission arrangement for the imaging optics 13 are chosen, where no location dependence of the width of the spectra on the CCD chip 14 consists. Alternatively, as in 3 shown, an arrangement in reflection possible. The imaging optics 13 then has a first microlens array 18 , a slit mask 28 for defining the entrance slit and a 45 ° to the beam direction 21 of the light arranged diffraction grating 29 as a focusing element. The grid 29 is designed to compensate for the optical transit times (path lengths) concave. The CCD chip 14 is 90 ° to the beam direction 21 arranged. In front of the CCD chip 14 is the aperture 17 arranged.

An evaluation unit 30 calculated from the on the CCD chip 14 registered spectra 25 the depth information about the xy coordinates on the sample surface 10 , For this purpose, the evaluation unit 30 a data processing unit, in a manner known per se (Fourier transformation), the further processing and evaluation of the entirety of the on the CCD chip 14 Performs individual spectra.

For example, with the in DE 103 42 175 A1 described surface profile resolved locally in the xy direction, this surface profile can be supplemented with the aid of the present invention by additional statements about the thickness of the sample to be examined at selected discrete points of the sample. Assuming a steady course of the thickness change, a three-dimensional model of the sample can also be calculated using the obtained measurement results.

Which discrete measuring points are involved is of minor importance for carrying out the measuring method. The position of the measuring points on the sample surface is freely selectable and is usually determined once in the development of the detector design as a function of the measurement task and adapted to the problem. For this, the selection of a suitable coordinate system takes place. In a simple case, as described above, the measuring points are determined by a rectangular grid (Cartesian coordinate system). But there are also applications conceivable in which advantageously an arrangement in radial pattern will be used according to a description of the measuring points in a polar coordinate system. Basically, the number and arrangement of the measuring points are selected according to the space available on the light sensor, on the one hand, and the desired resolution, on the other hand.

All features set forth in the description, the appended claims and the drawings may be considered individually. as well as in any combination with each other essential to the invention.

LIST OF REFERENCE NUMBERS

1

measuring device

2

FD-OCT arrangement

3

light source

4

Sample, cornea

5

measuring arm

6

reference arm

7

mirror

8th

detector

9

beamsplitter

10

sample surface

11

DOE

12

collimating optics

13

imaging optics

14

Light sensor, CCD chip

15

Microlens array

16

volume grating

17

cover

18

first microlens array

19

microlens

20

(free)

21

beam direction

22

second microlens array

23

beam

24

subaperture

25

spectrum

26

sensor surface

27

projection

28

slit mask

29

Concave, focusing diffraction grating

30

evaluation

Claims (5)

Detector (8) to which backscattered light from a sample (4) positioned in an OCT arrangement (2) is fed, for the simultaneous, two-dimensional spatially resolved registration of a plurality of spectrally resolved intensity distributions (25) of the backscattered light, each intensity distribution ( 25) is associated with a discrete point on the sample surface (10), the detector (8) having imaging optics (13) and a two-dimensional light sensor (14) onto which the intensity distributions (25) are imaged, characterized in that the imaging optics (13) comprises a microlens array (15) and a dispersive optical element in the form of a volume grating (16), wherein the volume grating (16) is a diffractive optical element in the form of a computer generated hologram. Detector (8) after Claim 1 , wherein the imaging optics (13) comprises a diaphragm (17) for masking the undiffracted portions of the light and each side of the symmetrical spectrum (25). Measuring device (1) for determining the thickness of a sample (4) at a plurality of discrete points with a single image, with an OCT arrangement (2) in which the sample (4) is positioned, with a detector (8) behind Claim 1 or 2 , and with an evaluation unit (30), designed to determine the depth information for these points from the spectral data thus obtained. Method for determining the thickness of a sample (4) at a plurality of discrete points with a single image, wherein the sample (4) is positioned in an OCT arrangement (2), backscattered light from the sample (4) is fed to a detector (8) which records spectrally resolved intensity distributions of the backscattered light associated with a plurality of discrete points on the sample surface (10) simultaneously and two-dimensionally spatially resolved, and Depth information for these points can be determined from the spectral data obtained in this way. Using a detector (8) according to Claim 1 or 2 and / or a measuring device (1) according to Claim 3 and / or a method according to Claim 4 for measuring the thickness of the cornea (4) of the human eye.

Images