WO2016054743A1

WO2016054743A1 - Retinal fundus imaging method and apparatus

Info

Publication number: WO2016054743A1
Application number: PCT/CA2015/051023
Authority: WO
Inventors: David Alexander Kahn
Original assignee: Annidis Corporation
Priority date: 2014-10-08
Filing date: 2015-10-08
Publication date: 2016-04-14

Abstract

A retinal fundus imaging apparatus and method employing Maxwellian illumination is described. The retinal illumination is focused at and passes through an illumination region within a working area, the working area being within and concentric with the circular ocular lens. The working area also includes a second, collection region through which the retinal reflection passes. The two regions (that is, the illumination region and the collection region) are separated and located on opposing sides of the working area so as to reduce or substantially remove unwanted lenticular backscatter from the collection path.

Description

Retinal Fundus Imaging Method and Apparatus

FIELD

[0001] The present disclosure relates generally to a method and apparatus for imaging the eye. More particularly, the present disclosure relates to a method and apparatus for imaging the retinal fundus.

BACKGROUND

[0002] The fundus of the eye, or retina, is a complex layered structure arranged in an approximately spherical shape at the back of the eyeball. It contains the light sensing rods and cones that enable vision. It is nourished by oxygenated blood supplied through arterioles and removed through venules. The nerve impulses from the rods and cones are directed to the brain through the optic nerve on the fundus, corresponding to the blind spot.

[0003] Direct visual observation of the retinal fundus can be accomplished using an ophthalmoscope, an instrument that has been around in various forms for over 150 years. The ophthalmoscope employs a light source, means for coupling the light into the eye through the pupil, and means for collecting light reflected back from the fundus and presenting an image of the fundus to the observer. The eye responds to continuous light by constricting the pupil size and so reducing the amount of light available to form an image. For this reason, the eye pupil may have to be chemically dilated using a mydriatic.

[0004] A fundus camera is similar to the ophthalmoscope but provides a permanent record of the fundus image in the form of a photograph. It also enables the use of a short, powerful flash of light to replace the continuous light required for the ophthalmoscope, and so sometimes avoiding the need for a mydriatic. The fundus camera uses an electronic image sensor such as a charge-coupled device (CCD) and the image is stored electronically. It may be displayed on a monitor or printed out as a photograph.

[0005] The fundus camera is used to examine and provide a record of the retina of the human eye in order to assess its state of health and to explore for incipient disease or other disorder. Illumination is normally introduced through the pupil and light reflected from the retina forms an image that also passes through the pupil. The two optical paths (that is, the illumination path and the reflected light or the collection path) are in close proximity to allow them both to pass through the pupil. However, they are not collinear and are externally fully separated to direct light from the illumination source to the retina (the illumination path) and to direct the retinal image to an image sensor (the reflected light path or the collection path).

[0006] The fundus image is dominated by the appearance of the optic nerve and the vascular structure of arterioles and venules. It is substantially of the color red, this coming from the blood, with some regions having an orange or yellow bias. The ophthalmologist is able to use this visual image to aid in the diagnosis of the health of the eye. Thorough diagnosis requires the use of a battery of other oculometric instruments in addition to the fundus camera.

[0007] Several factors commonly limit the quality of the image obtainable from a fundus imager. A prime demographic for eye pathologies consists of older people, but the ocular lens of an older person often tends to scatter a proportion of light passing through it. Incident illuminating light then becomes partially backscattered from the lens and this light reduces the contrast of the retinal image. Moreover, older people generally have smaller pupils, increasing the difficulty of injecting sufficient illumination and collecting sufficient energy for a good quality image.

[0008] Therefore, improvements in fundus imaging methods and apparatuses are desirable.

SUMMARY

[0009] Other aspects and features of the present disclosure will become apparent to those ordinarily skilled in the art upon review of the following description of examples in conjunction with the accompanying figures.

[0010] According to a first aspect of the present disclosure there is provided a retinal fundus imaging apparatus and method employing Maxwellian illumination. The retinal illumination is focused at and passes through an illumination region within a working area, the working area being within the ocular lens. The working area also includes a second, collection region through which the retinal reflection passes to obtain the retinal image. The two regions (that is, the illumination region and the collection region) are separated and located on opposing sides of the working area so as to reduce or substantially remove unwanted lenticular backscatter from the collection path. The separation between the two regions may be maximized by locating the two regions in close proximity to the circular periphery of the working area on opposing sides of the circle describing the working area. The working area may be determined based on the pupil area and the retinal field of view. The axis of separation may take any orientation. The two regions are also referred to as active areas or regions in the present disclosure.

[0011] In an example related to the first aspect, the two regions may be separated horizontally.

[0012] In another example related to the first aspect, the two regions may be equal in area, irrespective of their shapes.

[0013] In another example related to the first aspect, the two regions are segments with the straight borders parallel.

[0014] In another example related to the first aspect, one region or active area may be circular while the other region or active area may have a crescent shape and the center of the circular area may coincide with the center of curvature of the inner border of the crescent.

[0015]

[0016] In another example related to the first aspect, the separation distance may be substantially equal to t*tan(0/2) where T is the thickness of the ocular lens and Θ is the angular field of the retinal image. The term "substantially" is to be understood as including deviations of up to 10%. In a second aspect of the present disclosure, the working area may be divided into a plurality of regions, including an illumination region and a collection region. The division of the working area may be implemented using a spatial filter at a conjugate image location where the filter has two apertures one for transmitting the light illuminating the retina and the other for transmitting the light reflected from the retina. The remainder of the working area corresponds to the region where the filter may be opaque. The filter plane may be normal to the optical axis. The illuminating light may include infrared radiation.

[0017] In an example related to the second aspect, an illuminating light path and a reflection light path may be fully separated using a discrete mirror associated with one of the illumination or the reflection light paths.

[0018] In another example related to the second aspect, an illuminating light path and a reflection light path may be fully separated using a flexible optical fiber lightguide to form the illumination path prior to the spatial filter, where the light-bearing area of the lightguide corresponds to the area of the illumination region and the lightguide is bent away from the collection path.

[0019] In an example related to the second aspect, the spatial filter may be selected based on ocular characteristics of a patient and/or on measurement characteristics. The ocular characteristics of the patient include pupil size and astigmatism and the measurement characteristics include multi-spectral imaging, FAF, stereo imaging, and oxy-deoxy measurements.

[0020] In an example related to the second aspect, a plane mirror may be deployed adjacent to one of the apertures at a non-normal angle to the optical axis such that all the light passing through that aperture also reflects off the mirror but none of the light passing through the other aperture is so intercepted.

[0021] In another example related to the second aspect, the mirror may be movable to be associated with either aperture.

[0022] In another example related to the second aspect, the plane of the mirror may be at 45 degrees to the optical axis causing a deflection in the associated light path of 90 degrees.

[0023] In another example related to the second aspect, the opaque area of the spatial filter may be highly absorptive.

[0024] In another example related to the second aspect, the two regions or active areas may have transmission and reflection characteristics that have assigned spectral profiles.

[0025] In an example related to the second aspect, the assigned spectral profiles may comprise a low pass edge filter and a high pass edge filter.

[0026] In an example related to the second aspect, the two regions or active areas may have transmission and reflection characteristics that have assigned polarization characteristics.

[0027] In an example related to the second aspect, the assigned polarization characteristics may be those of polarizers having orthogonal polarization orientations.

[0028] In another example related to the second aspect, the spatial filter may be one of a set, each of which is designed for a specific range of pupil sizes and for a specific type of image based on diffuse reflection or autofluorescence or specular reflection or retinal contour observation, or for a patient with astigmatism, and where any one of the set can be selected and placed in position during the course of a retinal examination.

[0029] In a third aspect, the working area division may be implemented using a spatial filter combined with a coplanar mirror at a conjugate image location where the plane of the combination may be angled to the central optical axis leading to the eye so as to separate the illumination path from the reflection path and may contain an active area or region that reflects light and an active area or region that transmits light. The light may include infrared radiation.

[0030] In an example related to the third aspect, the plane of the combination may have an angle of 45 degrees to the optical axis. [0031] In another example related to the third aspect, the shape of the regions or the active areas defined on the combination may be those shapes defined for the active areas or regions within the working area projected according to the angle of the plane of the combination and magnified according to any magnification between the working area and its conjugate image.

[0032] In another example related to the third aspect, the combination area not assigned for transmission or reflection may be coated so as to be absorptive.

[0033] In another example related to the third aspect, the two active areas or regions may have transmission and reflection characteristics that have assigned spectral profiles.

[0034] In another example related to the third aspect, the assigned spectral profiles may comprise a low pass edge filter and a high pass edge filter.

[0035] In another example related to the third aspect, the two active areas or regions may have transmission and reflection characteristics that have assigned polarization characteristics.

[0036] In another example related to the third aspect, the assigned polarization characteristics may be those of polarizers having orthogonal polarization orientations.

[0037] In another example related to the third aspect, the combination may be one of a set, each of which is designed for a specific range of pupil sizes and for a specific type of image based on diffuse reflection or autofluorescence or specular reflection or retinal contour observation, or for a patient with astigmatism, and where any one of the set can be selected and placed in position during the course of a retinal examination.

[0038] In accordance with yet another aspect of the present disclosure, there is provided an apparatus for imaging a retina of an eye through an ocular lens employing Maxwellian illumination. The apparatus comprises: a spatial filter defining a first aperture and a second aperture, the first aperture and the second aperture for defining a working area within the ocular lens, the first aperture for transmitting light received from a light source; an optical lens arrangement for receiving light from the first aperture and for guiding the light received from the first aperture through an illumination region of the working area to illuminate the retina, the optical lens arrangement also for receiving light reflected from the retina from a collection region of the working area and for guiding the light received from the collection region towards the second aperture of the spatial filter; and an imaging system for receiving light transmitted through the second aperture of the spatial filter, the first aperture, the second aperture, and the lens arrangement being configured to have the illumination region spaced apart from the collection region, the illumination region being on a side of the working area, the collection region being on an opposite side of the working area.

BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Embodiments of the present disclosure will now be described, by way of example only, with reference to the attached Figures.

[0040] FIGURE 1 shows a conventional ocular lens working area arrangement with concentric illumination and collection areas.

[0041] FIGURE 2 shows a retinal fundus imaging method employing Maxwellian illumination in accordance with an aspect of the present disclosure.

[0042] FIGURE 3 shows retinal illumination and retinal image collection paths to a retinal location on the optical axis of the eyeball in accordance with an aspect of the present disclosure.

[0043] FIGURE 4 shows a circle-with-crescent arrangement of the working area assuming a pupil diameter of 5 mm, full path separation, and an angular field of 45 degrees.

[0044] FIGURE 5 illustrates lens backscatter geometry.

[0045] FIGURES 6A and 6B show a double segment arrangement of the working area assuming a pupil diameter of 5 mm and 3.5 mm, respectively, full path separation, and an angular field of 45 degrees.

[0046] FIGURE 7 shows a double circle arrangement of the working area assuming a pupil diameter of 5 mm, full path separation, and an angular field of 45 degrees.

[0047] FIGURE 8 shows an illumination path with combined spatial filter and mirror in accordance with an aspect of the present disclosure.

[0048] FIGURE 9 shows an illumination path with separated spatial filter and mirror in accordance with an aspect of the present disclosure.

[0049] Figure 10 shows an illumination path with a spatial filter and a contiguous optical fiber illumination path in accordance with an aspect of the present disclosure.

DETAILED DESCRIPTION

[0050] Generally, the present disclosure provided for a method and apparatus for imaging the retinal fundus.

[0051] According to an aspect of the present disclosure, there is provided a retinal fundus imaging apparatus and method employing Maxwellian illumination. As shown in Figure 2, the method 200 comprises defining a working area within the ocular lens (202). The retinal illumination is focused at and passes through an illumination region within the working area (204). The working area also includes a second, collection region through which the retinal reflection passes to obtain the retinal image (206). The two regions (that is, the illumination region and the collection region) are separated and located on opposing sides of the working area so as to reduce or substantially remove unwanted lenticular backscatter from the collection path. The separation between the two regions may be maximized by locating the two regions in close proximity to the circular periphery of the working area on opposing sides of the circle describing the working area. The axis of separation may take any orientation.

[0052] Maxwellian Arrangement and Gullstrand Separation

[0053] The illumination design of a conventional fundus camera follows the Maxwellian arrangement wherein a conjugate image of the illumination source is positioned within the ocular lens. With this arrangement, the illuminated volume of the lens is minimized thus reducing the backscatter. The backscatter image degradation mechanism is further mitigated by a design technique attributed to the Swedish ophthalmologist, Allvar Gullstrand. The Gullstrand design separates the optical illumination and collection paths within the lens by presenting illumination through an annular area within the ocular lens (the lens within the eye), while the retinal image is collected through a concentric circular virtual pupil within the ocular lens as shown in Figure 1 . This arrangement is implemented using a beamsplitter comprised of a combination of a spatial filter with a coplanar mirror that is located to coincide with a conjugate image plane of the center of the ocular lens.

[0054] In the conventional design, the spatial filter consists of a mirror surface with an aperture in the center such that the annular illumination reflects from the mirror part while the retinal reflection is collected through the central aperture. Where the beamsplitter is at 45 degrees to the optical paths, the aperture must take an elliptical form. The complementary design with illumination introduced through the center and the image collected through the annulus is also a possibility. The central illumination design is compatible with the efficient use of small LEDs and allows topological details of the retinal surface to be identified if the image sensor had a non-uniform polar response. It also has a more graceful degradation relationship to small pupils. It further has a better performance in the presence of cataracts. However, the axial path makes the central illumination design susceptible to the presence of unwanted reflections especially from lOL's. Also the peripheral collection results in a small depth of focus making focusing more critical and sometimes preventing some parts of the retina from being in good focus.

[0055] A partial separation of the two optical paths will mitigate the ocular backscatter but some backscatter will remain if the two optical paths are not fully separated. The amount of backscatter depends on the angular field of illumination and upon the separation distance between the central aperture and the inner boundary of the illuminating annulus. The backscatter will be reduced as the angular field is reduced and as the aperture to annulus separation is increased. However, both of these steps incur penalties. It is generally not desirable to create retinal images with small angular field coverage; the common standard is to acquire an angular field of 45 degrees or more. A small angular field leaves much of the retina unseen.

[0056] Moreover, the outer diameter of the illumination annulus is limited by the pupil diameter of the patient. Therefore, as the aperture to annulus separation is increased, the annulus becomes thinner and ultimately disappears, reducing and ultimately removing the illuminating energy. Similarly, if the separation is increased by reducing the aperture size, the collected energy of the retinal image reduces, ultimately to zero.

[0057] Typically, the design of the fundus imager is a compromise between the need for sufficient collected retinal image energy, the need to reduce the amount of lenticular backscatter, and the desirability of operating with small pupils and so avoid a need for mydriasis.

[0058] The requirement to suppress lenticular backscatter can be eased if the eye is dilated causing the pupil diameter to expand to perhaps 6 mm. Such dilation is standard. However both patients and health care professionals prefer to avoid mydriasis as this has time, cost and inconvenience impacts. Yet another limitation is the need to avoid illuminating the iris, a possibility when the size of the annulus is increased. An illuminated iris becomes another source of unwanted light that degrades the retinal image. Ideally, there should be some margin to reduce this probability of occurrence.

[0059] Where mydriasis is not used, there is inevitably a minimum pupil size below which the image quality becomes unsatisfactory because of iris and lenticular backscatter and insufficient energy in the retinal image. Moreover, if the design is optimized for use with a small pupil, it will be less than optimal when used with a large pupil.

[0060] A further problem with conventional fundus camera design is exacerbated when capturing retinal images of patients with cataracts. The cataract normally is located in the center of the ocular lens and so obstructs the optical path used for retinal image collection. As cataracts are also characteristic of the older demographic that is in most need of eye care, this poses yet another limitation on the utility of conventional fundus cameras.

[0061] The Working Area

[0062] The light to and from the retina must be confined within a circular working area concentric with the ocular lens, where the diameter of the working area is set by the pupil diameter and the angular field associated with the required fundus image. Typically, the working area diameter is less than the pupil diameter by t*tan(6/2) where "t" is the thickness of the ocular lens and "Θ" is the full angular fundus field. For a typical value for T of 4 mm, and an angular field of 45 degrees, the working area diameter is 1 .7 mm less than the pupil diameter.

[0063] To achieve full separation of the illumination and image collection paths, the separation distance between the border of the virtual collection pupil within the working area and the border of the illumination region or area within the working area must exceed t*tan(6/2) typically taking a value of 1 .7 mm as shown above.

[0064] Where the two areas associated with illumination and image collection are formed by a central circle and a peripheral annulus, a total of 2t*tan(6/2) of the working area diameter would be taken up with separation, corresponding to a value of 3.3 mm using the typical values used earlier. This would imply that a pupil diameter of 5 mm, corresponding to a working area diameter of 3.3 mm would leave no area for illumination and collection if full path separation is required.

[0065] Consequently, full path separation can only be achieved with large pupils, e.g., of 6 mm diameter. To obtain such large pupils commonly requires mydriasis. Without mydriasis, commonly only partial path separation can be achieved.

[0066] Retinal Fundus Imaging

[0067] Aspects of the present disclosure rely on advantageously dividing the working area into a plurality of regions, including at least one collection region and at least one illumination region. In particular, it is recognized that the division of the working area into a central circle and a peripheral annulus requires the working area diameter to include two path separation distances, one on each side of the central area, a requirement that substantially erodes the proportion of the working area that can be dedicated to illumination and collection. [0068] Accordingly, in an aspect of the present disclosure, the working area is divided such that the illumination region or area is located near the periphery of the working area on one side and the collection region or area is similarly located but on the opposite side as shown in Figure 3. Such an arrangement requires that the path separation distance only factors once into the working area diameter. This separation could take any orientation including the horizontal and vertical. Similarly, there is an unlimited variety of shapes that could be assigned to the two areas. For ease of reference, example orientations and shapes will be described. It is noted that the present disclosure is not to be limited to the example orientations and shapes. As the central part of the working area would not be used, this arrangement is well suited for eyes with cataracts where the center of the lens is commonly the most opaque.

[0069] In another aspect of the present disclosure, the spatial filter characteristics are adjustable to better match the working area of the patient's eye. Thus a patient with a small pupil and by extension a small ocular working area will be illuminated with a smaller area of illumination and a patient with a large pupil will be illuminated with a larger field of illumination, enabling a good quality image to be obtained from a range of pupil sizes. Such adjustment may take the form of having a multiplicity of spatial filters, one of which could be selected for use as best matched to the pupil size of the patient.

[0070] A feature of using horizontal separation is the compatibility with capturing an image suitable for contributing towards a stereo presentation. Such a presentation would require the capture of two such images, one being generated by illuminating on the left side of the working area and collecting on the right side, while the other would be generated by illuminating on the right side and collecting on the left side.

[0071] To maximize the strength of the image for a given source radiance, the areas of illumination and image collection may be equalized and could take the shape of opposing segments whose straight borders are separated. If this would produce an excessive strength such as may saturate the image sensor or cause undue patient discomfort, the illumination area may be reduced by increasing the separation further and reducing the segment size.

Alternatively, the illuminating shape could be reduced to that of a small circular area; this has the advantageous feature of producing images that contain clearly discernable features generated by the surface contours of the retina if the image sensor has a non-uniform polar response. [0072] The image collection area may also be made circular; this has the advantage of making the image quality more uniform. A small circular area, while collecting less illumination, also results in less distortion from any eye lens astigmatism and in a greater depth of focus. To maximize the illumination area when used with a circular collection area, the shape of the illumination area should be a crescent where the radius of the inner border is greater than the radius of the collection area by the path separation distance. See Figure 4 which is described in further detail later.

[0073] While various combinations of shapes have been discussed, it should be appreciated that any shape may be used that conforms to the requirement that the illumination and collection areas be on opposite sides of the working area, lying substantially within opposing semicircles.

[0074] Moreover, while with a 45 degree angular field it is in principle possible to achieve full path separation with pupils having a diameter in excess of about 3.3 mm, it remains valid that for smaller pupils only partial path separation is possible. However, using aspects of the present disclosure, a compromise between the need to collect sufficient image energy and reject unwanted backscatter yields a better result than possible with conventional fundus cameras where with the same angular field the minimum pupil diameter for full path separation was about 5 mm. This is because a diameter across the working area of the conventional fundus camera must include two path separation distances, one either side of the center. On the other hand, a diameter across the working area in the apparatus according to the present disclosure requires only one path separation distance, resulting in more area being available for illumination and collection for a given path separation distance. The lens backscatter geometry is illustrated in Figure 5.

[0075] With a lesser angular field, the required pupil diameter would be less; for example, with a 30 degree angular field, the pupil diameter for full path separation would need to be in excess of 2.1 mm. In all cases, the excess required above the minimum would be determined by the need to capture sufficient energy for the image, a requirement in turn limited by the radiance of the source and the duration of the illuminating pulse.

[0076] Figure 6A shows a double segment arrangement of the working area assuming a pupil diameter of 5 mm, full path separation, and an angular field of 45 degrees. The separation shown is 1 .7 mm in the horizontal plane and equal areas are assigned to illumination and collection. The working area diameter would be 3.3 mm and each segment would have a maximum thickness of 0.8 mm and an area of about 1 .7 square millimeters. Figure 6B shows a similar arrangement for a pupil diameter of 3.5 mm and full path separation. In order to achieve this while maintaining the same illumination and collection areas, the angular field would be reduced to 23 degrees accompanied by a path separation of 0.8 mm.

[0077] Figure 7 shows a double circle arrangement of the working area assuming a pupil diameter of 5 mm, full path separation, and an angular field of 45 degrees. The separation shown is 1 .7 mm in the horizontal plane and equal areas are assigned to illumination and collection. The working area diameter would be 3.3 mm and each circle would have a diameter of 0.8 mm and an area of about 0.55 square millimeters.

[0078] As described earlier, Figure 4 shows a circle-with-crescent arrangement of the working area assuming a pupil diameter of 5 mm, full path separation, and an angular field of 45 degrees. The separation shown is 1 .7 mm in the horizontal plane and equal areas are assigned to illumination and collection. The working area diameter would be 3.3 mm. The circle would have a diameter of 1 .2 mm and an area of about 1 .2 square millimeters while the crescent would have a maximum thickness of 0.45 mm and the same area.

[0079] The working area may be divided using a spatial filter combined with a beamsplitter as shown in Figure 8. This component is co-located with a conjugate image of the working area plane. It may consist of a plane at 45 degrees to the central optical axes, having areas that are ideally fully reflective, fully transmissive or fully absorptive. The absorptive area is used to help suppress any unwanted light. If the beamsplitter is at 45 degrees, the partitions defined for the working area must be projected at 45 degrees to form the shapes of the corresponding areas on the beamsplitter; for example, a circle in the working area would become an ellipse in the beamsplitter. Other angles may be used and the projections modified accordingly. This arrangement is used in conventional fundus cameras where a central elliptical hole is surrounded by a reflecting surface.

[0080] An unwanted feature of using a spatial filter having a plane oriented at an angle other than normal to the optical axis is that the conjugate image of the ocular lens working area, being in a plane normal to the optical axis, only partially coincides with the spatial filter. This results in a sub-optimum performance of the spatial filter. Therefore, it may be advantageous to split the functions of spatial filter and mirror and have them provided by separate devices as shown in Figure 9. In this case, the spatial filter will be wholly located at the locus of the conjugate image of the working area. The spatial filter may consist of a thin metal mask with two apertures, one for the illumination path and one for the collection path, the entire mask being coated with an absorptive layer to absorb any light not required for illumination and any stray light that could degrade the image. The mirror may be set at 45 degrees or some other convenient angle to the optical axis and may have a plane reflective surface that intercepts nominally one side of the illumination beam. It may be located beside the spatial filter such that the incident illumination would first be reflected and then filtered. Typically, the intercepted proportion would correspond to less than 50% of the working area conjugate image.

[0081] Figure 10 shows the eye on the left, two lenses L1 and L2 in the centre and the spatial filter on the right. A fiber lightguide is shown abutting the spatial filter and being bent away from the main optical axis towards the illumination source.

[0082] The spatial filter is located at a conjugate image location of the ocular lens. An effect of the two lenses is to image the ocular lens on to the spatial filter with a magnification equal to the ratio of the lens focal lengths, that is f2/f1 .

[0083] Light from the illuminating fiber is collected by L2 and substantially collimated. The collimated light passes to L1 where it is focused down to an image within the lens of the eye. The light then passes on to and illuminates the retina.

[0084] Light reflected at the retina is first collected by the ocular lens that substantially collimates the image and passes it to L1 . The reflected light then creates a virtual image at an intermediate point between the lenses located at a distance f2 from the lens L2. This light is then collimated again by L2 and passed to the spatial filter that is located at a distance f2 from the lens L2. The light passes through the spatial filter and is directed to the image sensor that includes further lens elements (not shown) to create the image.

[0085] Where the eye is unable to accommodate, i.e. in the case of short or long sightedness, the system compensates by adjusting the spacing between L1 and L2 to ensure that the virtual image of the retina remains at a constant distance f2 from the lens L2.

[0086] As an alternative to using a mirror in association with a separate spatial filter to separate the illumination and collection paths, the illumination can be provided through a flexible optical fiber lightguide that terminates at the spatial filter and bends away from the illumination path.

[0087] Where a multiplicity of different illumination sources are required, particularly at different wavelengths, each illumination source may be assigned a different fiber and each fiber directed to the spatial filter as described above. [0088] Where illumination sources may vary greatly in their strength (optical power), the effective area of the associated lightguide could be selected to be smaller or larger to accommodate the differences and ensure a suitable amount of optical power passes into the eye.

[0089] Where a fiber is used to transport the illumination power to the spatial filter, all the surplus power is removed before the illumination is launched into the fiber, substantially removing the need to take careful measures to remove surplus light that could enter into the collection path and degrade the image.

[0090] To allow for the creation of stereo images, the mirror may be set on a sliding mechanism allowing it to be positioned to intercept either half of the illumination path. A first image of the stereo pair may be collected with the mirror in one position and the second image of the stereo pair may be collected with the mirror in the other position.

[0091] Advantageously, the proportion of the illumination that is not to be reflected at the mirror may be masked off before the illumination is introduced to the mirror. Where the mirror is mobile, the prior blocking mask is arranged to move with it. This prevents the ingress of high light levels into the optical chamber containing the path to the image sensor. However, such ingress would require measures to suppress unwanted light and to prevent it from reaching the image sensor. Ideally, all unwanted light must be absorbed. However, a single surface rarely absorbs more than 90% of the incident light and scatters/reflects the remainder. This reflected light must therefore be somehow captured and absorbed. However, when it is diffusely reflected (scattered), it propagates in all directions and capture/absorption is non-trivial. It is preferable to block and capture the unwanted illumination before it approaches near to the collection path and any unwanted scattering then has a very low probability of reaching the image sensor.

[0092] Alternatively, the mirror may be positioned to be associated with the aperture collecting the light reflected from the eye. This arrangement provides for easier suppression of the incident illumination not required for the eye and that could reach the image sensor and degrade the image.

[0093] It may be practically advantageous to apply some magnification between the working area and its conjugate image as this reduces the required strength of the optical re-focusing. It also changes the angular and linear positioning tolerances applying to the spatial filter and mirror or integrated beamplitter. For example, assuming an angular field of 45 degrees, with a magnification of x2, the linear positioning tolerance of the spatial filter will be doubled while the angular positioning tolerance of the mirror will be reduced by 30%. These changes ease the physical design where any one of a multiplicity of spatial filters with different partitioning characteristics to match the characteristics of the patient's eye and the specific requirements of the examination, must be placed in position.

[0094] The multiplicity of spatial filters is required to provide good imaging from a multiplicity of patient pupil sizes. However, other spatial filter characteristics can usefully be provided within the multiplicity. In particular, either or both the illumination and collection areas may be given spectral profiles such that they preferentially block or transmit certain spectral bands. This could be used to provide autofluorescence images of the retina where illumination is provided at one wavelength and the image captured using a longer wavelength. For fluorescence imaging, there is no requirement to separate the illumination and collection paths within the ocular lens as any unwanted backscatter is removed by ancillary spectral filters.

[0095] Dichroic (spectrally selective) surfaces could also be used to create a stereo image where the wavelengths are close enough to provide a similar image but far apart enough to be separated by the beamsplitter. In rapid succession, one spectral band of illumination may be reflected on one side and the reflection transmitted on the other side, followed by a second spectral band of illumination that would be reflected on the other side but transmitted on the first side, thus creating a stereo pair. This would require one side to be equipped with a high pass edge filter while the other side is equipped with a low pass edge filter.

[0096] A further type of spatial filter could employ polarization elements that would enable polarization sensitive aspects of the retinal image to be either identified and measured or suppressed. Suppression would usefully be applied to unwanted specular reflection components.

[0097] A typical beamsplitter/spatial filter combination would measure about 10 x 15 mm and consist of a glass substrate of thickness 2 mm, coated as needed with a metallic layer for the reflective area, an anti-reflection coating for the transmissive area and a black coating for the absorptive area.

[0098] A typical spatial filter would measure about 10 x 10 mm and consist of a thin metal substrate of thickness 0.2 mm entirely black coated and with two or more shaped apertures for collection and illumination. Where a mirror is used to separate the illumination and collection paths, the mirror would typically be a plane of size 5 x 15 mm; this could be located between two black absorptive plane elements of the same size, allowing the moveable mirror to direct light to either the right or left of the spatial filter while blocking illumination on the other side. Where a fiber is used to separate the illumination and collection paths, it would abut the spatial filter. Where a multiplicity of fibers are used, they would each abut the spatial filter. The shape and size of the fiber should match the shape and size of the aperture of the spatial filter. The fiber lightguide may be formed by a single fiber, typically circular, or a multiplicity of small fibers bundled to form the required shape.

[0099] In the preceding description, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the embodiments. However, it will be apparent to one skilled in the art that these specific details are not required. Also, features from various examples may be combined in a single embodiment, as appropriate.

[00100] The above-described embodiments are intended to be examples only. Alterations, modifications and variations can be effected to the particular embodiments by those of skill in the art. The scope of the claims should not be limited by the particular embodiments set forth herein, but should be construed in a manner consistent with the specification as a whole.

Claims

WHAT IS CLAIMED IS:

1 . A retinal fundus imaging method to image a retina of an eye through an ocular lens employing Maxwellian illumination, the method comprising:

defining a working area within the ocular lens;

focusing a retinal illumination at and passing through an illumination region within the working area;

collecting a retinal reflection passing through a collection region within the working area to obtain an image of the retina, the illumination region and the collection region being separated and located on opposing sides of the working area so as to reduce or substantially remove unwanted lenticular backscatter within the retinal image; and

dividing the working area into the illumination region and the collection region using a spatial filter at a conjugate image location where the spatial filter has a first aperture for transmitting light illuminating the retina and a second aperture for transmitting the light reflected from the retina.

2. The method of claim 1 , wherein the separation between the illumination region and the collection region is maximized by locating the two regions proximate to a circular periphery of the working area on opposing sides of a circle describing the working area.

3. The method of claim 1 , wherein the working area is concentric to the ocular lens.

4. The method of claim 1 , wherein the working area is determined based on at least one of a pupil size and a retinal field of view.

5. The method of claim 1 , wherein an illuminating light path and a reflection light path are separated using a discrete mirror associated with one of the illumination path or the reflection light path.

6. The method of claim 1 , further comprising selecting the spatial filter based on at least one of ocular characteristics of a patient and measurement characteristics.

7. The method of claim 6, wherein the ocular characteristics of the patient include pupil size and astigmatism and, the measurement characteristics include multi-spectral imaging, FAF, stereo imaging, and oxy-deoxy measurements.

8. A retinal fundus imaging apparatus configured to implement the method of any one of claim 1 to 7.

9. An apparatus for imaging a retina of an eye through an ocular lens employing

Maxwellian illumination, the apparatus comprising:

a spatial filter defining a first aperture and a second aperture, the first aperture and the second aperture for defining a working area within the ocular lens, the first aperture for transmitting light received from a light source;

an optical lens arrangement for receiving light from the first aperture and for guiding the light received from the first aperture through an illumination region of the working area to illuminate the retina, the optical lens arrangement also for receiving light reflected from the retina from a collection region of the working area and for guiding the light received from the collection region towards the second aperture of the spatial filter; and

an imaging system for receiving light transmitted through the second aperture of the spatial filter, the first aperture, the second aperture, and the lens arrangement being configured to have the illumination region spaced apart from the collection region, the illumination region being on a side of the working area, the collection region being on an opposite side of the working area.