Open Access
Add to BibSonomyAbstract
The current augmented-view device for glaucoma patients are limited by the low angle minification (less than 3X), image overlapping and pupil mismatching. We present a novel ultra-thin near-eye augmented-view device (UNAD) with a proper minification angle of 4X to avoid scanning eye movements. The device is realized by one symmetrical ultra-thin off-axis eight-mirror reversed telescopic optical system (field-of-view (FOV): x:-34°~3°; y: −5°~-20°, 5°~20°) and one central un-minified optical system (FOV: −5°~5°), and hence, the system can achieve a FOV (x:-34°~3°; y: −20°~20°) without overlapping images. Furthermore, the device fully accounts for pupil matching with the entrance pupil of the human eye. Finally, the simulation results in CODE V and LightTools verify that our proposed system is feasible and has potential practical value for glaucoma patients.
© 2016 Optical Society of America
1. Introduction
Glaucoma, the second-leading cause of blindness worldwide after cataracts [1], is characterized by high intraocular pressure, optical nerve atrophy, and visual field loss. Glaucoma patients have the symptoms of vision loss gradually over a long period, as the peripheral vision decreases, leading to difficulties in spotting obstacles, avoiding obstacles, noticing friends on the street, etc [2]. In the final stages, the remaining central vision (tunnel field) is only 10 degrees [3]. To overcome this drawback, many augmented-view devices for glaucoma patients have been developed.
Reverse telescopes [4] are a traditional device to expand the visual field. However, the minification is not very high, usually less than 3X, because higher minification results in longer overall physical length. Therefore, most of these devices are not commercially available. To solve the minification limitation of the traditional reverse telescope, Vargas-Martin [5] et.al combined a see-through head-mounted display with a simultaneous visual field expander to enlarge the patients’ field-of-view (FOV) by 3–7 times. However, this approach is only realized in image processing techniques, without much concern for the actual optical system, which has a serious problem with image overlapping. Furthermore, this device loses the color and depth information. Additionally, Hoeft [6] et.al presented that the central vision resolution may be reduced as the FOV is increased, and proposed an amorphic lens reverse telescope to minify only the horizontal meridian, which would reduce the impact to the central resolution, but at a cost of severe image distortion.
Another approach to expand the visual field for patients with tunnel vision is using prisms. The principle for considering the use of prisms is based on field-shifting. Two apex-to-apex prisms are placed in front of one of the patient’s eyes and provide a laterally shifted visual field, while a conventional spectacle lens is placed in front of the other eye to ensure the original residual visual field [7] is retained. The patients will get a combination of the original visual field and the shifted visual field to capture more information about their environment. However, the images of two different objects in the scene may overlap on the retinas, which may cause visual confusion [7].
Apart from these visual field expanders for patients with normal central vision, some devices for age-related macular degeneration (AMD), marked by people who have a low central vision, have also been proposed. Tremblay [8, 9] et.al presented a telescopic contact lens with two independent optical paths for AMD. The magnified optical path incorporates an arrangement of annular concentric reflectors to achieve 2.8X magnification of the angle, while light passing through a clear central aperture retains un-magnified normal vision. However, the light propagating through these two optical paths does not utilized the entire pupil of the human eye, which may result in missing parts of the image.
Being inspired by the above current work, this paper innovatively proposes an ultra-thin near-eye augmented-view device (UNAD) for patients with a tunnel field of 10°, which achieves a 4x minification angle, eliminates image overlapping, and matches the entire pupil of the human eye. This device would render the minification factor to a proper value to prevent scanning eye movements over a wide visual field [5], and guarantee the high resolution for the human eye. With higher minification factors, the image becomes too small and difficult to distinguish. Considering different observation requirements, we designed the system with two independent optical paths; one is a symmetrical ultra-thin off-axis eight-mirror reversed telescopic optical path to expand the peripheral field, and the other supplies normal central vision without any angle minification to guarantee the central vision resolution [6]. With this structure, patients with tunnel vision can obtain a large peripheral FOV without influencing their central FOV. The two independent optical paths will be switched by selectively blocking the central and peripheral aperture. Therefore, when patients gaze at near objects, they can choose the minified vision mode to get zoom-out images, which ensures wide-field peripheral vision, but the device can be switched to normal vision for looking at far objects. The theory for pupil matching and the overall system design are described in Section 2. Then the simulation results are presented in Section 3.
2. Design of ultra-thin near-eye augment-view device (UAND)
The dissatisfaction of the existing augmented-view devices for glaucoma arises from several causes, for example low angle minification (less than 3X), image overlapping, and pupil mismatching. We innovated the UNAD, which achieves a good angle minification (up to 4X), eliminates image overlapping, and matches the entrance pupil of the human eye. The details of the method are illustrated through the following subsections.
2.1 Pupil matching of the exit pupil of the UNAD with the entrance pupil of the human eye
The switchable telescopic contact lens proposed by Tremblay [8] et.al has not considered the pupil matching problem, which may result in loss of all the light rays from one of the fields during eye movement [9], as shown in Fig. 1(a),(b). In Fig. 1(a), the red rays represent normal vision, and Depd2 is the corresponding exit pupil diameter, while the blue and green rays indicate the top and bottom of the vision, respectively. Their corresponding exit pupil diameters are Depd1 and Depd3, while D represents the diameter of the human eye pupil. Since pupil matching was not considered, the resulting exit pupil diameters of the different independent optical paths are different from the entrance diameter of the human eye pupil. All the light rays in Fig. 1 (a) can successfully pass through the mismatched pupil before eye movement; however, the entire green light ray field misses after the eye moves, which can result in missing parts of the image, as shown in Fig. 1(b).
Fig. 1 The light propagation at the pupil of the human eye by a mismatched device pupil and a matched device pupil during eye movement. (a) Light rays pass through the mismatched pupil before eye movement, (b) Light rays pass through the mismatched pupil after eye movement, (c) Light rays propagate through the matched pupil before eye movement, (d) Light rays propagate through the matched pupil after eye movement, and (e) Human eye during eye movement.
Our ultra-thin near-eye reverse telescopic lens is different from Tremblay’s lens in two aspects. The first difference is that our system is used to minify angle for glaucoma patients with tunnel vision while Tremblay’s device is aimed to magnify angle for patients of age-related macular degeneration (AMD). The second difference is that our system fully accounts for matching the pupil along different independent optical paths, as shown in Fig. 1(c),(d). The exit pupil diameters for the different optical paths and the entrance pupil diameter of the human eye have the following Eq.:
Fig. 1(c) shows the light ray propagation conditions by matching the pupil before eye movement, and the Fig. 1(d) presents the light tracing path after matching the pupil after eye movement resulting in vignetting of the field. While this may only affect the luminance of the image, it can be solved by expanding the UNAD system’s exit pupil diameter.
2.2 Overall system design
Based on the pupil constraints depicted in Section 2.1, the UNAD has been designed. Table 1 lists the specifications of the optical system. As shown in Table 1, the FOV of the system consists of three independent parts, which are the top FOV (x:-34°~3°, y: −5°~-20°), the central FOV (x:-5°~5°, y: −5°~5°) and the bottom FOV (x:-34°~3°, y: 5°~20°). Among these three FOV parts, the top FOV and the bottom FOV are symmetrical and used for the minified optical paths, while the central FOV is for the un-minified normal optical path.
Table 1. Specifications of the optical system
The optical system, as demonstrated in Fig. 2, consists of one symmetrical ultra-thin off-axis eight-mirror reverse telescopic optical system and one normal central optical system with 1X minification, which are independent of each other. The ultra-thin reverse telescopic lens is designed to narrow the large FOV to a range that is within the range of the patients’ tunnel vision. The central region is utilized for normal vision without any angle minification to ensure central resolution. The minified optical path corresponds to the top FOV and bottom FOV, while the central optical path corresponds to the central FOV. The two independent optical paths work with the help of orthogonal polarization films [10], where the un-minified and minified vision will switch by selectively blocking the central and peripheral aperture, respectively. With this setup, the system can separate the images of the different FOVs on the retina solving the problem of image overlapping, simultaneously taking into consideration the matching of the device’s pupil with human eye’s entrance pupil.
Fig. 2 Optical layout of the UNAD lens. (a) Un-minified (1X) optical path through the central clear aperture, (b) Minified (1/4X) multiple-reflection optical path, and (c) Expanded view.
Fig. 2 shows our ultra-thin near-eye augmented-view lens design. The symmetrical ultra-thin off-axis eight-mirror reverse telescopic lens is 34.6 mm in diameter, while the normal central lens aperture is 8.29 mm in diameter, and the whole system is 10 mm thick. The distance between the last surface of the device and the human eye pupil is set to 28 mm to ensure that the device meets the requirements for different people. The exit pupil diameter of the overall system is 2 mm matching with the diameter of human eye pupil. Furthermore, a polarizer combined with a pair of switching liquid crystal (LC) glasses are placed in front of the device to selectively block the central aperture or the peripheral aperture, and we make the central two reflectors semi-transparent, which can assure no obstructions with light rays propagating through the central vision.
As given in Fig. 2, the single off-axis ultra-thin near-eye reverse telescopic lens consists of 4 reflectors, which can dramatically reduce the size and weight and result in an achromatic design. We set the surfaces of the four mirrors to XY polynomials to correct the astigmatism, coma, and other higher-order aberrations, which are easily brought in by the wide-field system. The Eq. for the XY Polynomial surface is given in Eq. (2).
where is the sag of the surface parallel to the z-axis, is the curvature at the vertex, is a conic constant, is the coefficient of the monomial.To get an easily aligned configuration, the entire optical system should be processed from a single piece of glass made from Poly methyl Meth acrylate (PMMA). With that configuration, the difficulty in alignment can be greatly reduced compared to a freeform system.
3. Simulation results
According to the system in Section 2, the UNAD with a FOV(x:-34°~3°; y: −20°~20°) has been designed for patients with glaucoma. In order to correctly simulate the system of the human eye with the new device, we designed a Liou and Brennan Eye model [11], which is widely accepted as an accurate and comprehensive eye model. It accounts for the cornea, aqueous humor, offset pupil, and curved retina surface. The model we adopted included the aspheric coefficient. The Liou eye model is shown in Fig. 3(a).
Fig. 3 (a)The Liou eye model, (b)(c) Polychromatic MTF curves of the central optical path (with eye model), and (d)(e) Polychromatic MTF curves of the peripheral optical path (with eye model).
In our research, we placed the Liou optometry model at the exit pupil plane to assist in evaluating the imaging Euality of the new augmented-view device. Fig. 3(b),(c) shows the polychromatic modulation transfer function (MTF) for the central optical path (including the optical model of the eye) and Fig. 3(d),(e) shows the polychromatic modulation transfer function (MTF) for the peripheral optical path (including the optical model of the eye). It indicates that the MTF values for the central vision are close to the diffraction limit, while the MTF values of the peripheral vision are all above 0.5 at 50 cycles/mm at all the fields.
The distortion grid of the central system is given in Fig. 4(a), which shows that it is undistorted. Fig. 4(b) gives the distortion grid for the peripheral vision. As shown in Fig. 4(b), the maximum geometric distortion is less than 10%, and the real image grid achieves good image quality except that the whole image grid is offset to the right and bottom slightly. The RMS wave-front error of the peripheral system is shown in Fig. 4(c), and its value is below λ/50 at 546 nm wavelength. These results show that the system performs very well.
Fig. 4 (a) The distortion grid of the central optical path, (b) The distortion grid of the peripheral optical path, and (c) The RMS wave-front error of the peripheral optical path.
Fig. 5 shows that, if the UNAD has its exit pupil planes located at the entrance pupil of the human eye, and matched with the size of the entrance pupil diameter, the light rays from different fields will diverge from each other on the retina. Fig. 5(a) shows the footprint of the central un-minified vision (semi - central FOV (x:-5°~5°; y: −5°~0°)) on the retina, and Fig. 5(b) gives the footprint of the peripheral minified vision (only the top FOV) on the retina. The whole central un-minified vision consists of two symmetrical semi-central FOVs and the whole peripheral minified vision consists of the top FOV and the bottom FOV, which are symmetrical. The two independent optical paths (the central and peripheral path) work with the help of orthogonal polarization films, where the un-minified and minified vision will be switched by selectively blocking the central and peripheral aperture, respectively.
Fig. 5 Footprint plot of central normal vision and peripheral minified vision. (a) Footprint plot of central normal vision, and (b) Footprint plot of peripheral minified vision.
The vision of the ultra-thin near-eye augmented-view was modeled using LightTools with the parameters listed in Table 2. As shown in Table 2, the same object (height: 30mm, width: 16mm, object distance: 500mm) is located at seven different positions corresponding to the three independent FOVs. Fig. 6 shows the results of the LightTools Simulation. As given in Fig. 6(a), we use seven objects, which cover the whole FOV of the UNAD system, as the sampling points to evaluate the imaging quality of the system. Figure 6(b) shows light rays from the same object at three different positions pass through our system and enter into the eye model through its entrance pupil. Fig. 7 shows ray-tracing results of the seven objects, and the FOVs of the traced objects’ center point are (−34°, −20°), (0°, −20°) (3°, −20°), (0°, 0°), (−34°, 20°), (0°, 20°) and (3°, 20°). As shown in Fig. 7, the display images of top FOV and bottom FOV on the retina are minified, and the image of central FOV is unchanged. The luminance uniformity of the central image is about 80%, while the peripheral region’s luminance uniformity is beyond 95%. The average luminance value of the central region is about four times that of the peripheral ones’, and the average luminance contrast is about 0.5, which meets requirements for human eye use [12]. Furthermore, the image of the top FOV (corresponding to the bottom imaging region) has little displacement to the right and bottom from the ideal image region; the image of bottom FOV offsets to the right and top only a little for the UNAD is symmetric, both of which are caused by system distortions discussed in Section 3.
Table 2. Specifications of the object
Fig. 6 LightTools simulation. (a) The distribution of the object in the UNAD, and (b) The light ray path layout.
Fig. 7 Simulation results and illuminance chart. (a) The simulation result at three different positions; (b) The illuminance chart of the optical path
6. Conclusion
Due to the dissatisfaction of the existing augmented-view devices for glaucoma based on parameters such as low angle minification, image overlapping, pupil mismatching, etc., this study presents a novel UNAD for patients with a tunnel field of 10°. Considering different observation requirements, the UNAD consists of one symmetrical ultra-thin off-axis eight-mirror reverse telescopic optical system and one central un-minified optical system, which are independent from each other. The two independent optical paths can be switched by changing orthogonal polarization films. However, improvements could be made to the UNAD to remove system distortions. In addition, the UNAD has the potential of being redesigned as a rotationally symmetric system, which may have additional practical applications.
Acknowledgments
This work is supported by the National Natural Science Foundation of China (NSFC) (Grant No. 61471039).
References and links
1. S. Kingman,“Glaucoma is second leading cause of blindness globally,” Bull. World Health Org. 82(11), 887–888 (2004).
2. D. S. Friedman, E. Freeman, B. Munoz, H. D. Jampel, and S. K. West, “Glaucoma and mobility performance: the salisbury eye evaluation project,” Ophthalmology 114(12), 2232–2237 (2007).
3. A. H. Zalta, “Use of a central 10 degrees field and size V stimulus to evaluate and monitor small central islands of vision in end stage glaucoma,” Br. J. Ophthalmol. 75(3), 151–154 (1991). [CrossRef] [PubMed]
4. N. Drasdo, “Visual field expanders,” Am. J. Optom. Physiol. Opt. 53(9), 464–467 (1976). [CrossRef] [PubMed]
5. F. Vargas-Martín and E. Peli, “Augmented-view for restricted visual field: multiple device implementations,” Optom. Vis. Sci. 79(11), 715–723 (2002). [CrossRef] [PubMed]
6. W. W. Hoeft, W. Feinbloom, R. Brilliant, R. Gordon, C. Hollander, J. Newman, E. Novak, B. Rosenthal, and E. Voss, “Amorphic lenses: a mobility aid for patients with retinitis pigmentosa,” Am. J. Optom. Physiol. Opt. 62(2), 142–148 (1985). [CrossRef] [PubMed]
7. R. L. Woods, R. G. Giorgi, E. L. Berson, and E. Peli, “Extended wearing trial of Trifield lens device for ‘tunnel vision’,” Ophthalmic Physiol. Opt. 30(3), 240–252 (2010). [CrossRef] [PubMed]
8. E. J. Tremblay, I. Stamenov, R. D. Beer, A. Arianpour, and J. E. Ford, “Switchable telescopic contact lens,” Opt. Express 21(13), 15980–15986 (2013). [CrossRef] [PubMed]
9. R. E. Fischer and B. Tadic-Galeb, Optical System Design (McGraw-Hill, 2008), pp. 29–34.
10. A. Arianpour, G. M. Schuster, E. J. Tremblay, I. Stamenov, A. Groisman, J. Legerton, W. Meyers, G. A. Amigo, and J. E. Ford, “Wearable telescopic contact lens,” Appl. Opt. 54(24), 7195–7204 (2015). [CrossRef] [PubMed]
11. H. L. Liou and N. A. Brennan, “Anatomically accurate, finite model eye for optical modeling,” J. Opt. Soc. Am. A 14(8), 1684–1695 (1997). [CrossRef] [PubMed]
12. M. W. Cannon Jr., “Perceived contrast in the fovea and periphery,” J. Opt. Soc. Am. A 2(10), 1760–1768 (1985). [CrossRef] [PubMed]
