WO2013083147A1

WO2013083147A1 - Focusing lens and system for optical coherence tomography

Info

Publication number: WO2013083147A1
Application number: PCT/EP2011/006232
Authority: WO
Inventors: Claudia Gorschboth; Tobias Jeglorz; Ole Massow; Klaus Vogler; Henning WISWEH
Original assignee: Wavelight Gmbh
Priority date: 2011-12-09
Filing date: 2011-12-09
Publication date: 2013-06-13

Abstract

A focusing lens 26 exhibits such imaging properties that a ray bundle 22a, 22b incident on the object side is focused on the image side to a focus 58a, 58b in such a manner that a principal ray 56a, 56b of the ray bundle 22a, 22b runs on the image side parallel to an optical axis 40 of the focusing lens 26 and with a lateral spacing A from the optical axis 40 that is variable in a manner depending on an object-side angle of incidence θ of the ray bundle 22a, 22b, and that an axial position B of the focus 58a, 58b is displaced in the axial direction z away from the focusing lens 26 with increasing object-side angle of incidence θ of the ray bundle 22a, 22b.

Description

Focusing lens and system for optical coherence tomography

The invention relates to a focusing lens, in particular for a system for optical coherence tomography.

In optical coherence tomography (OCT for short) the use of, for example, achromats, flat-field scanning lenses, f-theta scanning lenses or telecentric lenses is known. A factor that is common to lenses of such a type is that they each generate an image-field surface that is concavely curved towards the lens or represents a plane that is oriented perpendicular to the optical axis of the lens.

However, a cornea to be surveyed with an OCT system, for example, is not plane/flat but possesses a surface that is curved away from the OCT system or possesses an internal structural surface that is curved away from the OCT system. This has the consequence that when a lens known from the state of the art is employed in an OCT system the image-field surface is not two- dimensionally congruent with the curved boundary surface to be surveyed. As a result, it is impossible to generate a sharp focus two-dimensionally over the entire boundary surface. This results in deterioration of the OCT signal in the region of the boundary surface, where the examining ray impinges on the boundary surface not in focus but out of focus.

It is therefore an object of embodiments of the invention to optimise a focusing lens for an OCT system that is employed for the purpose of imaging the human eye. By virtue of an optimised beam control over the entire scan field, the lateral resolution and the depth of focus should be improved, in order thereby to enable the recording of a tomogram of an object to be examined having high image quality and a good signal-to-noise ratio (SNR for short).

According to embodiments, a focusing lens is provided with such imaging properties that a ray bundle incident on the object side is focused on the image side to a focus (at least one focus) in such a manner that a principal ray of the ray bundle runs on the image side parallel to an optical axis of the focusing lens and with a lateral spacing from the optical axis that is variable in a manner depending on an object-side angle of incidence of the ray bundle, and that an axial position of the focus is displaced in the axial direction away from the focusing lens with increasing object-side angle of incidence of the ray bundle.

The focusing lens according to the invention consequently makes it possible that a ray bundle entering the focusing lens obliquely (i.e. non-axially) in relation to the optical axis of the focusing lens is focused on the image side to a focus, the lateral spacing of which from the optical axis of the focusing lens and the axial spacing of which from the focusing lens are each greater than the corresponding lateral and axial spacings of a focus to which a ray bundle entering the focusing lens axially is focused on the image side.

As a result, the focusing lens can be utilised to scan a surface of an object to be examined that is vaulted away from the focusing lens. In particular, it is possible that the beam waists of all the ray bundles passing through the focusing lens come to be situated on or in the curved surface of the object, even though the surface of the object is vaulted away from the focusing lens, i.e. is convexly curved, viewed from the focusing lens. Since the lateral resolution of an OCT system depends on the lateral beam diameter of the ray bundles, in both ray bundles the surface of the object is accordingly observed with the maximal lateral resolution. Consequently, neither in the region of the focus of an axial ray bundle nor in the region of the focus of a ray bundle entering the focusing lens obliquely do losses of imaging performance occur. Rather, the imaging performance of the focusing lens is made uniform over the entire field coverage.

The imaging properties of the focusing lens according to the invention

consequently have a positive effect on the entire imaging performance of an OCT system equipped with this focusing lens. The focusing lens makes it possible to improve the lateral resolution and the depth of focus and to generate tomograms having a high image quality and a good signal-to-noise ratio.

In addition, the focusing lens according to the invention enables a telecentric imaging on the image side, since the principal rays of the ray bundles run on the image side, independently of the angle of incidence, parallel to the optical axis of the focusing lens. This has the advantage that the point of incidence and the incident angle ε of the principal ray are known. The incident angle ε is needed in order to be able to calculate the further course of the beam, since the beam direction changes in accordance with Snell's law of refraction, with

sin(£)-n = sinfc -n'. In this equation, ε is the incident angle or angle of incidence of the principal ray in relation to the surface normal of a refracting surface, ε' is the angle of reflection of the principal ray in relation to the surface normal of the refracting surface, n is the refractive index of the medium upstream of the refracting surface, and n' is the refractive index of the medium downstream of the refracting surface. This so-called refractive correction is needed for the purpose of reconstructing the tomogram. In addition, axial displacements of the object to be examined in relation to the focusing lens do not result in a change in the observed or illuminated image-field surface.

Rather, the image-field surface is substantially constant in relation to such axial displacements.

The focusing lens further causes the principal ray of a ray bundle to run with a lateral spacing A from the optical axis that changes in a manner depending on the object-side angle of incidence Θ of the principal ray. This means that the lateral spacing, that is to say, the spacing perpendicular to the optical axis, is a non-constant function depending on the angle of incidence Θ: A = Α(θ).

The focusing lens preferentially exhibits such imaging properties that the function Α(θ) is continuous in Θ. In particular, it is conceivable that the focusing lens has such imaging properties that the function Α(θ) is linear in Θ, i.e. Α(θ)/ Θ = constant. For example, the focusing lens has such imaging properties that Α(θ) = f-θ, where f is a constant length. For example, f is the focal distance of an axial ray bundle, i.e. the spacing of the focus of a ray bundle entering the focusing lens at an angle of incidence Θ = 0° from the nearest optical surface of the focusing lens. Consequently f may also represent the focal length of the focusing lens for the axial ray bundle.

The axial position of the focus, that is to say, the position of the focus along the optical axis of the focusing lens, can be combined with an axial spacing B. The axial spacing B is the spacing of the focus with respect to the plane that intersects the optical axis of the focusing lens perpendicularly at the focus of the ray bundle that has entered the focusing lens axially. In other words: by this definition, the axial spacing B of the axial ray bundle is zero, i.e. Β(θ=0^ο) = 0, whereas all the axial spacings of the ray bundles entering the focusing lens obliquely (i.e. modulo[e,360°]≠0°) are greater than zero, i.e. Β(θ≠0^ο) > 0.

The focusing lens preferentially exhibits such imaging properties that the axial spacing B is a non-constant function depending on the angle of incidence Θ, that is to say, B = Β(θ). In particular, it is conceivable that the focusing lens has such imaging properties that the function Β(θ) is non-linear in Θ, i.e. Β(θ) is not directly proportional to Θ. For example, the focusing lens has such imaging properties that the function Β(θ) depends in non-linear manner on the lateral spacing Α(θ), that is to say, B = Β(θ) = B(A) = Β(Α(Θ)). The focusing lens may have such imaging properties that for all angles of incidence Θ the spacing B and the spacing A satisfy the following condition:

where a and b are each a real, positive constant having the unit of a length. Additionally the following may hold: 0 < |A|≤ a and/or 0 < |B| < b. The image-field surface of the focusing lens is accordingly curved in accordance with a partial portion of the surface of an ellipsoid of revolution, the axis of symmetry of which coincides with the optical axis of the focusing lens. The parameters a and b then represent the lengths of the two different semi-axes of the ellipsoid of revolution. For example, the following may also hold: a = b. The image-field surface then exhibits a spherically curved central portion with radius of curvature a (=b). Values for a and/or b are, for example, 6.0 mm, 6.8 mm, 7.7 mm, 8.0 mm and/or 11.0 mm.

The focusing lens is preferentially configured in such a manner that the lateral position Α(θ) and the axial position Β(θ) of the focus are displaced with the angle of incidence Θ in such a manner that the focus position [Α(θ), Β(θ)] describes a curved image-field surface with increasing object-side angle of incidence Θ. For example, the curved image-field surface may be vaulted away from the focusing lens. In this case, in comparison with the vaulting of the image-field surface of a conventional achromatic lens it is therefore possible to speak of an inverse vaulting of the image-field surface.

In a particularly preferred embodiment the image-field surface of the focusing lens is congruent with a predetermined surface, curved in dish-like or dished manner, on or in the object to be examined. The curved surface may represent a boundary surface on which an optical index of refraction changes.

Alternatively, however, in the case of the surface it may also be a question of a surface situated on the object or within the object that runs parallel, that is to say, with a constant spacing with respect to a curved boundary surface of the object. So if the angle of incidence of a ray bundle entering the focusing lens is tuned continuously, on the image side the ray bundle is focused continuously onto the curved surface being observed. The beam waist of the ray bundle then always comes to be situated, independently of the angle of incidence, on the curved surface on or in the object. In particular in this connection, the refraction of light of the ray bundles on the boundary surfaces of the object, which the ray bundle passes through after leaving the focusing lens right up to its image-side focus, can also be taken into account.

For example, the curved image-field surface may have been adapted to a curved surface on or in a human eye. In particular, it is conceivable that the curved image-field surface has been adapted to a curved surface on or in the cornea and/or on or in the human lens.

Furthermore, it is conceivable to use the focusing lens also for the purpose of recording projections onto the cornea, the lens and/or other curved surfaces of the object.

The focusing lens may further exhibit the property that the optical path lengths through which ray bundles of varying angles of incidence pass from a common object-side intersection point right through the focusing lens as far as the image-side focus positions thereof are independent of the angle of incidence at which the ray bundles enter the focusing lens. Accordingly, the optical path lengths passed through by the different ray bundles over the entire available scan field are equally long. The optical path length can be kept constant over the entire scan field. For this purpose the focusing lens may comprise a multi- lens system, the components of which have differing refractive indices.

The focusing lens may comprise one or more optical components. The optical components may have differing optical refractive indices. At least one optical surface of the focusing lens may exhibit an aspherical free-form surface. The imaging properties of the focusing lens are preferentially brought about by the shaping of at least one free-form surface.

The focusing lens may have been formed in rotationally symmetrical manner with respect to the optical axis.

A system for optical coherence tomography comprises, in addition to a focusing lens described above, a light-source for making available a ray bundle of coherent light, controllable scanning components for routing the ray bundle emitted from the light-source through the focusing lens at a variable object-side angle of incidence, and a control device that has been set up to control the scanning components with a view to acquiring tomograms of an object to be examined.

Merely for the sake of completeness, at this point the following terms will be clarified: the image-field surface is that surface onto which the totality of image- side foci of a large number of ray bundles entering the focusing lens at various angles of incidence is imaged.

If it is assumed that a z-axis along the optical axis of the focusing lens extends from the focusing lens in the direction of the image-side image-field surface and a tangential surface T touches the image-field surface tangentially at an arbitrary point P on the image-field surface, a convexly vaulted/curved image- field surface, viewed from the focusing lens, is distinguished by the fact that the z-coordinate of the intersection point at which the optical axis intersects the tangential surface T exhibits a smaller z-value than the z-value of the z- coordinate of point P. The principal ray is the ray of a ray bundle that runs from an object point though the midpoint of the aperture stop of the optical system. The principal ray enters the optical system along a straight line that is directed towards the midpoint of the entrance pupil, and leaves the system along a straight line that runs through the midpoint of the entrance pupil. The principal ray, which is combined with a conical ray bundle which emanates from a point on a scanning component, behaves effectively like the middle ray of the ray bundle and represents it.

On the image side' relates to the space-half situated beyond the focusing lens, in which the object to be examined (the specimen) has to be arranged for the purpose of creating a tomogram, whereas 'on the object side' relates to the space-half situated on this side of the focusing lens.

The optical axis of the focusing lens is the connecting-line of the centres of curvature of the optically refracting surfaces of the focusing lens. If the focusing lens is rotationally symmetrical, the optical axis coincides with the axis of symmetry.

The invention will be elucidated further in the following on the basis of the appended drawings, in which:

Fig. 1 shows a schematic overall representation of a system for optical coherence tomography according to one embodiment.

Figs. 2a to 2d each show a schematic representation of a focusing lens

pertaining to the state of the art,

Fig. 3 shows a schematic representation of a human eye and also

some parameters of the geometry thereof,

Fig. 4 shows a further schematic representation of the focusing lens from Fig. 2d, Fig. 5 shows a schematic representation of a further focusing lens pertaining to the state of the art,

Fig. 6 shows a schematic representation of a focusing lens according to one embodiment, and

Fig. 7 shows a further schematic representation of the focusing lens from Fig. 6 and also some parameters of the focusing lens.

Appended furthermore to the description are Tables, of which

Tab. A contains some parameters relating to the focusing lenses shown in Figs. 2a to 2d, and

Tab. B contains some parameters relating to the focusing lens shown in

Fig. 7.

A system for optical coherence tomography is denoted generally in Fig. 1 by 10. In the exemplary case the system 10 serves for examining an object shown in the form of a human eye 12. Optical coherence tomography is based, for example, on so-called time-domain (TD for short) OCT or on so-called

frequency-domain (FD for short) OCT.

The system 10 includes a light-source 14 for emitting coherent light. The light- source 14 is designed, for example, for the purpose of FD OCT as a tuneable light-source or emits a spectrum of coherent light that is broadband within the frequency space.

The light emitted from the light-source 14 is directed onto a beam-splitter 16. The beam-splitter 16 is a constituent part of an interferometer 18 and splits up the incident optical output in accordance with a predetermined splitting ratio, for example 50:50. One ray bundle 20 runs within a reference arm; another ray bundle 22 runs within a specimen arm. Instead of the free-space setup represented in Fig. 1, the interferometer 18 may also have been realised partly or entirely with the aid of fibre-optic components. The light that has been branched off in the reference arm impinges on a mirror 24 which reflects the light back onto the beam-splitter 16 collinearly. For the purpose of TD OCT the mirror 24 may be displaceable along the direction of propagation of the ray bundle 20. The light that has been branched off in the specimen arm impinges on the object 12 to be examined, which back-scatters or reflects back the light in the direction of the beam-splitter 16.

In Fig. 1 a three-dimensional Cartesian coordinate system x, y, z of the system 10 has been drawn in schematically. In this connection the z-axis represents the direction of observation for examining the object 12. This coordinate system is also used in Figs. 2a to 7.

Within the specimen arm a focusing lens 26 and controllable scanning

components 28 are provided. The controllable scanning components 28 have been set up to route the ray bundle 22 coming in from the beam-splitter 16 through the focusing lens 26 onto the object 12. In this connection the angle of incidence at which the ray bundle 22 coming from the beam-splitter 16 enters the focusing lens 26 is adjustable with the aid of the scanning components 28. In the example shown in Fig. 1 the scanning components 28 take the form of rotatably supported mirrors for this purpose. The axes of rotation of the mirrors may be perpendicular to one another. The angle of rotation of the mirrors is, for example, set with the aid of an element operating in accordance with the galvanometer principle.

The focusing lens 26 focuses the ray bundle 22 onto or into the object 12. The focusing lens 26 is represented in Fig. 1 schematically as one lens but may comprise a plurality of different lenses, and will be described more precisely further below with reference to Figs. 2 to 7.

The light back-scattered from the object 12 in the specimen arm is collinearly superimposed at the beam-splitter 16 with the light reflected back from the mirror 24 in the reference arm so as to form an interference beam 30. The optical path lengths in the reference arm and specimen arm are substantially equally long, so that the interference beam 30 displays an interference between the ray bundles 20, 22 back-scattered from reference arm and specimen arm. A detector 32 registers the intensity of the interference beam 30 as a function of the time, the wavelength and/or the wave number. For this purpose the detector 32 may take the form of a photodiode or spectrometer.

The signal registered by the detector 32 is transferred to a control device 34. On the basis of the registered intensity of the interference beam 30 the control device 34 ascertains an interferogram via the wavelength or via the wave number, which serves as basis for tomograms of the object 12. The control device 34 controls the scanning motion of the scanning components 28 in such a manner that the acquisition of ID, 2D and/or 3D tomograms is possible. The ascertained tomograms are displayed on a display unit 36 and can be stored in a memory 38.

In Figs. 2a to 2d four embodiments, known in the state of the art, of the focusing lens 26 shown in Fig. 1 are represented more precisely. In addition, two exemplary ray bundles 22a and 22b have been drawn in. The principal ray of ray bundle 22a enters the focusing lens 26 axially along the optical axis 40 of the focusing lens 26. The angle of incidence Θ of ray bundle 22a accordingly amounts to 0°. On the other hand, the principal ray of ray bundle 22b enters the respective focusing lens 26 at an angle of incidence Θ different from zero.

In Fig. 2a a simple achromatic lens is represented. This focusing lens 26 focuses ray bundle 22a onto the paraxial image plane 42. The paraxial image plane 42 intersects the optical axis 40 of the focusing lens 26 at the focus of the axial ray bundle 22a and extends perpendicularly with respect to the optical axis 40. In contrast, the focusing lens 26 focuses ray bundle 22b onto an image-field surface 44 that is vaulted towards the focusing lens 26. The image-field surface 44 is that surface onto which the totality of image-side foci of a plurality of ray bundles 22, 22a, 22b which are incident into the focusing lens 26 at various angles of incidence is imaged. The image-field surface 44 is represented in dashed manner in Fig. 2a and is curved away from the paraxial image plane 42. This effect is known as Petzval field curvature. In Fig. 2b a flat-field scanning lens is represented. This focusing lens 26 focuses both the axially incident ray bundle 22a and the obliquely incident ray bundle 22b onto the paraxial image plane 42. The image-field surface 44 accordingly corresponds to the paraxial image plane 42. The lateral spacing A of the focus of the obliquely incident ray bundle 22b from the optical axis 40 of the focusing lens 26 amounts to f-tan(9), where f is the focal length of the focusing lens 26.

In Fig. 2c a basic f-theta scanning lens is represented. This focusing lens 26 likewise focuses both the axially incident ray bundle 22a and the obliquely incident ray bundle 22b onto the paraxial image plane 42. Accordingly, here too the image-field surface 44 corresponds to the paraxial image plane 42.

However, the f-theta scanning lens has such imaging properties that the lateral spacing A of the focus amounts to f-θ, where f is again the focal length of the focusing lens 26.

Lastly, in Fig. 2d a telecentric f-theta scanning lens is represented. This focusing lens 26 focuses, as in the cases of Figs. 2b and 2c, both the axially incident ray bundle 22a and the obliquely incident ray bundle 22b onto the paraxial image plane 42. Accordingly, also in this case the image-field surface 44 corresponds to the paraxial image plane 42. However, the principal rays of the ray bundles 22a, 22b emerging from the telecentric f-theta scanning lens on the image side run parallel to the optical axis 40, to be specific, independently of the angle of incidence Θ of the incident ray bundle 22a, 22b. Like the simple f-theta scanning lens, the telecentric f-theta scanning lens has such imaging properties that the lateral spacing A of the focus amounts to f-0, where f is again the focal length of the focusing lens 26.

In Tab. A, parameters have been compiled in exemplary manner that

characterise the focusing lenses 26 shown in Figs. 2a to 2d. In the first column the angle of incidence or scan angle Θ of a ray bundle is plotted. The value of Θ is specified in the angular dimension (°). In the second column the angle of reflection or output angle φ of the ray bundle corresponding to the angle of incidence Θ is plotted. The value of φ is also specified in the angular dimension (°). In the third and fourth columns the focus diameter or spot diameter along the major semiaxis and along the minor semiaxis of an ellipse are plotted, within which the ray bundle intersects the paraxial image plane 42. In the fifth column the focus area or spot area of this ellipse is plotted. In the sixth column the averaged power density within the spot of light is plotted. Columns three to six represent normalised parameters and have each been mathematically

normalised to the corresponding value for Θ = 0°. The effective focal length (EFL for short) of the respective lens is represented in brackets.

A feature common to the focusing lenses 26 shown in Figs. 2a to 2d is that the image-field surface 44 is curved towards the focusing lens 26 (see Fig. 2a) or lies within the paraxial image plane (see Figs. 2b to 2d). But the object 12 to be examined with the system 10, as well as the internal structures thereof, such as, for example, the cornea 46 or the human lens 48, are not flat or plane but have a curved geometry.

As represented in Fig. 3, the cornea 46, for example, exhibits an anterior radius of curvature of R « 7.7 mm and a posterior radius of curvature of R « 6.8 mm. The human lens 48 exhibits, for example, an anterior radius of curvature of R* « 11 mm and a posterior radius of curvature of R* « -6.0 mm. Furthermore, in Fig. 3 typical refractive indices n of the cornea 46, of the chamber of the eye 50 and of the vitreous body 52 have been entered. In addition, along the optical axis 40' of the eye 12 typical lengths/thicknesses for the cornea 46, the chamber of the eye 50, the lens 48 and the vitreous body 52 are specified in mm. The diameter of the eye along y amounts to about 24 mm.

The flat image-field surface 44 of the focusing lenses 26 shown in Figs. 2b to 2d and the non-flat geometry of a curved surface 54 of the eye 12 shown in Fig. 3 may yield problems. In the case of a scan field shown in Fig. 4, the scan field has a lateral extent D. The ray bundles 22b running aside the optical axis 40 (=400 impinge on the curved surface 54 of the eye 12 in defocused manner. If the curved surface 54 corresponds by approximation to a spherical surface with radius R, at the edge of the scan field an axial distance Δζ between the image- field surface 44 and the curved surface 54 results of Δζ = R - (R² - D²/4)^1/2. For typical values of R « 8 mm and D » 14 mm, Δζ = 4 mm results. According to Gaussian optics, at a wavelength λ of the light in ray bundle 22b of λ = 800 μιη and with a numerical aperture NA of the focusing lens 26 of NA » 0.02 in relation to the beam diameter w(0) = w₀ of ray bundle 22b at the location of the focus this axial distance Δζ results roughly in a doubling of the beam diameter νν(Δζ) of ray bundle 22b at the location of incidence of ray bundle 22b on the curved surface 54, νν(Δ^ζ) = w₀ (1 + (A-Az)²/(n-w₀ ²)²)^1/2-

Similar (or even still more pronounced) broadenings of the beam diameter w(Az) arise also in the cases shown in Figs. 2a to 2c. This has the consequence that a focusing lens 26 shown in Figs. 2a to 2d does not generate a sharp focus over the entire scan field D or the entire object depth Δζ. The focus is sharp only for a distance (namely for Δζ = 0; see, e.g. Fig. 4). This results in deterioration of the signal, of the signal-to-noise ratio and of the lateral resolution.

In Fig. 5 a non-telecentrically designed focusing lens 26 is shown. In this case the ray bundles 22a, 22b are perpendicularly incident onto the curved surface 54. To realize the ray path shown in Fig. 5, a very much larger lateral diameter of the focusing lens 26 is necessary, making the focusing lens expensive and enlarging the entire optical system. In addition, the configuration with the ray bundles 22a, 22b running non-parallel to the optical axis 40 but convergently incident in the image-side region onto the curved surface 54 may make it difficult to correct OCT tomograms. In this configuration, the incident angle and the location of incidence onto the curved surface 54 are generally not known if the spacing between object 12 and focusing lens 26 without contact lens is variable, that is to say, the object 12 is not fixed. But the incident angle and the location of incidence are needed for calculating the transit-times of the ray bundles 22a, 22b within the interferometer 18, in particular for calculating the refractive correction in order to create correct OCT tomograms.

In Fig. 6 an embodiment of a focusing lens 26 according to the invention is shown schematically. In the case of Fig. 6 it is a question of a symbolic representation of the mode of operation of a focusing lens 26 according to the invention. The focusing lens 26 exhibits such imaging properties that ray bundle 22a that is incident (in particular, collimated, i.e. parallel) axially on the object side and ray bundle 22b that is incident (in particular collimated, i.e. parallel) on the object side at an angle of incidence different from zero in relation to the optical axis 40 of the focusing lens 26 are focused on the image side onto corresponding foci 58a, 58b in such a manner that a principal ray 56a of ray bundle 22a and a principal ray 56b of ray bundle 22b run on the image side parallel to the optical axis 40 of the focusing lens 26. The focusing lens 26 accordingly exhibits a beam path that is telecentric on the image side. The exit pupil of the focusing lens 26 lies at infinity on the image side.

The parallelism of the principal rays 56a and 56b with the optical axes 40, 40' is shown clearly in Fig. 6 with the aid of \\ symbols. In Fig. 6, in addition, the position at which the scanning component(s) 28 shown in Fig. 1 is/are located is denoted by 28.

The focusing lens 26 shown in Fig. 6 accordingly generates a telecentric image on the image side. This has the advantage that the point of incidence and the incident angle of the principal rays 56a, 56b on the object 12 do not change with varying spacing between object 12 and focusing lens 26. The image-side ray paths are consequently known and calculable. In comparison with the focusing lens 26 shown in Fig. 5 with convergent incidence of light onto the object 12, the aperture in the focusing lens 26 shown in Fig. 6 does not have to be so large.

The focusing lens 26 further exhibits such imaging properties that the ray bundles 22a and 22b are focused on the image side in such a manner that the principal rays 56a and 56b run with a lateral spacing A from the optical axis 40 that is variable in a manner depending on the object-side angle of incidence Θ of the ray bundle 22a, 22b. This means that the lateral spacing A, that is to say, the spacing perpendicular to the optical axis, is a non-constant function depending on the angle of incidence Θ, A = Α(θ). In the present example the focusing lens 26 has such imaging properties that Α(θ) = f-θ, where f represents the axial spacing of the focus 58a from the optical surface 59 of the focusing lens 26 coming closest to the focus 58a and/or the focal length of the focusing lens 26.

The focusing lens 26 additionally exhibits such imaging properties that the ray bundles 22a and 22b are focused on the image side onto the respective foci 58a and 58b in such a manner that an axial position B of the focus 58a or 58b is displaced away from the focusing lens 26 in the axial direction z with increasing object-side angle of incidence Θ of the ray bundle 22a or 22b.

This is illustrated in Fig. 6. The axial spacing B is the spacing of the focus 58b with respect to a plane 60 (which is represented in dashed manner in Fig. 6) that intersects the optical axis 40 perpendicularly at the focus 58a of a ray bundle 22a entering the focusing lens 26 axially. Except for the focus 58a of ray bundle 22a entering axially, all the foci 58b of the ray bundles 22b entering the focusing lens 26 obliquely are situated on the other side of the plane 60, viewed from the focusing lens 26.

In Fig. 6 the curved surface 54 of the object 12 is also denoted by 44. T is is intended to illustrate that the position of the focus 58a, 58b for a changing angle of incidence Θ follows the surface 54 which is curved in dish-like manner. With increasing lateral spacing A, the focus 58a, 58b is accordingly displaced further away from the focusing lens 26 in the positive z-direction. The axial spacing B is dependent on the lateral spacing A in such a manner that the image-field surface 44 is congruent with the curved surface 54 of the object 12 and is vaulted towards the object 12. So in relation to the focusing lens 26 from Fig. 2a the image-field surface 44 of the focusing lens 26 from Fig. 6 is vaulted inversely.

The optical properties of a focusing lens 26 shown symbolically/schematically in Fig. 6 are represented in exemplary manner in Fig. 7. The beam path of three exemplary ray bundles 22a, 22bl and 22b2 may, for example, be generated by one or more optical components 62a, 62b with a free-form surface 64 (which, in particular, is aspherical).

With a view to characterising the size of a focus, the so-called fraction of enclosed energy (FEE for short) is utilised. This quantity is a measure of the proportion of the entire energy transported by a ray bundle 22a, 22bl, 22b2 per unit of time that passes through a circular area with a given radius (radius from centroid; RFC for short). In Fig. 7, in the right-hand lower quadrant a diagram for the FEE is represented as a function of the RFC of a ray bundle 22a with a lateral spacing of 0 mm, of a ray bundle 22bl with a lateral spacing of 3.75 mm, and of a ray bundle 22b2 with a lateral spacing of 7.5 mm from the optical axis 40 at the location of the respective foci 58a, 58bl, 58b2 on the curved surface 54. In addition, a fourth curve DL has been drawn in which represents the diffraction limit of the focusing lens 26. It can be discerned well that all four curves substantially overlap. The foci 58a, 58b and 58b2 of the ray bundles 22a, 22bl and 22b2 are accordingly substantially equally large and diffraction- limited.

Further exemplary parameters charactering the focusing lens 26 from Fig. 6 can be gathered from Tab. B. In this connection, NA is the numerical aperture of the focusing lens 26. "Radius of curvature image field 8 mm" means that the image-field surface 44 exhibits a spherical radius of curvature Kr of 8 mm.

"Scan-field diameter 15 mm" means that the maximal lateral spacing S_F of two focal points of the image-field surface amounts to 15 mm. The focusing lens 26 comprises, in exemplary manner, two lenses, namely lens 1 and lens 2. Lens 1 exhibits two spherically curved boundary surfaces, the radii of curvature of which are represented by Rl' and R2'. Lens 2 exhibits a spherically curved boundary surface with radius of curvature Rl (=R2 , and a free-form boundary surface, the curvature of which is characterised with the aid of the so-called sagitta z(r). In this connection r is a radial (i.e. lateral) spacing from the optical axis 40 and serves as variable for z(r), whereas c, R2, k and βι to β₈ are constant coefficients/parameters for z(r). The thickness d' (d) corresponds to the spacing of the two boundary surfaces of lens 1 (lens 2) along the optical axis 40.

The focusing lens 26 may further exhibit the property that the optical path length is substantially constant over the entire scan field. By virtue of the focusing lens 26 of multi-lens design, the elements 62a, 62b of which have differing refractive indices, the optical path length can be kept constant over the scan field.

The focusing lens 26 shown in Fig. 6 consequently enables the optimised recording of a tomogram of an anterior segment of the eye 12 (anterior- segment OCT). The optimisation consists, in particular, in the adaptation of the focus location in the beam direction to the curvature of the object 12 or of another arbitrarily shaped boundary surface 54 in the object 12, such as, for instance, the anterior surface of the human lens 48. For this purpose an arbitrarily curved image-field surface 44 is generated within the object 12. In particular in this connection, the refraction of light of the ray bundles 22a, 22b on the marginal surfaces of the tissue of the object 12 is also taken into account.

The focusing lens 26 shown in Fig. 6 may further exhibit the property that the optical path lengths that ray bundles 22, 22b or the principal rays 56a, 56b of the ray bundles 22a, 22b with differing angles of incidence Θ pass through from a common object-side intersection point 28 right through the focusing lens 26 as far as the image-side focus positions 58a, 58b thereof are independent of the angle of incidence Θ at which the ray bundles 22a, 22b or the principal rays 56a, 56b of the ray bundles 22a, 22b enter the focusing lens 26. Accordingly, the optical path lengths passed through by the differing ray bundles 22a, 22b or by the principal rays 56a, 56b of the ray bundles 22a, 22b are equally long over the entire available scan field of the focusing lens 26. Accordingly, the optical path length can be kept constant over the entire field. For this purpose the focusing lens 26 may comprise a multi-lens system (represented, for instance, in Fig 7), the components of which have differing refractive indices.

Unless expressly stated otherwise, identical reference symbols in the Figures stand for identical or identically-acting elements. In other respects, an arbitrary combination of the features elucidated in the Figures in connection with individual embodiments is conceivable.

Claims

1. Focusing lens (26) with such imaging properties that a ray bundle (22; 22a, 22b) incident on the object side is focused on the image side to a focus (58a, 58b) in such a manner that a principal ray (56a, 56b) of the ray bundle (22a, 22b) runs on the image side parallel to an optical axis (40) of the focusing lens (26) and with a lateral spacing (A) from the optical axis (40) that is variable in a manner depending on an object-side angle of incidence (Θ) of the ray bundle (22a, 22b), and that an axial position (B) of the focus (58a, 58b) is displaced in the axial direction (z) away from the focusing lens (26) with increasing object- side angle of incidence (Θ) of the ray bundle (22a, 22b).

2. Focusing lens (26) according to Claim 1, wherein the axial position (B) of the focus (58a, 58b) is displaced with the angle of incidence (Θ) in such a manner that the focal position (A, B) describes an image-field surface (44) that is curved in dish-like manner.

3. Focusing lens (26) according to Claim 2, wherein the image-field surface (44) is convexly vaulted, viewed from the focusing lens (26).

4. Focusing lens (26) according to Claim 2 or 3, wherein the image-field surface (44) has been adapted to a marginal surface (54) of an eye structure, in particular of a human cornea or lens.

5. Focusing lens (26) according to one of Claims 2 to 4, wherein the image- field surface (44) exhibits a curved central portion with a radius of curvature between 6 mm and 11 mm, the central portion optionally being spherically curved.

6. Focusing lens (26) according to one of the preceding claims, wherein the focusing lens (26) includes at least one component (62a, 62b) with an aspherical free-form surface (64).

7. Focusing lens (26) according to one of the preceding claims, wherein the focusing lens (26) further exhibits such properties that the optical path lengths through which ray bundles (26a, 26b) of varying angles of incidence (Θ) pass from a common object-side intersection point (28) right through the focusing lens (26) as far as the image-side focus positions (58a, 58b) thereof are independent of the angle of incidence (Θ) at which the ray bundles (26a, 26b) enter the focusing lens (26).

8. System (10) for optical coherence tomography (OCT), comprising:

a focusing lens (26) according to one of Claims 1 to 7,

a light-source (14) for making available a ray bundle (22; 22a, 22b) of coherent light,

controllable scanning components (28) for routing the ray bundle (22; 22a, 22b) emitted from the light-source (14) through the focusing lens (26) at a variable object-side angle of incidence (Θ), and a control device (24) that has been set up to control the scanning components (28) with a view to acquiring tomograms of an object (12) to be examined.