SMST Window Volume: I, Issue: 1, November 2008

Essay

Multi-Resolution Image Processing in selecting the Most Effective Visual Information for Retinal Prosthesis Debdoot Sheet and Manjunatha M. School of Medical Science and Technology, Indian Institute of Technology, Kharagpur, INDIA

Abstract— Each time we open our eyes, millions of photoreceptors convert light into neural signals and pass them to our inner nuclear layers, and ultimately our optic-nerves, via ganglion cells. In visually impaired patients suffering with Outer Retinal Degenerations (ORD) like Age-Related Macular Degeneration (AMD) and Retinitis Pigmentosa (RP), however, the photoreceptor cells cease to function although the inner layers of the retina are intact. Over the last decade, in their quest to partially restore visual perception in patients suffering from AMD or RP, researchers have been exploring the possibility of stimulating the function of photoreceptors by implanting electrodes inside the retina to generate neural signals. In this paper, we introduce an algorithm for visual prosthesis device, where the data processing complexity is dramatically reduced in order to correspond to the technological constraints of implantable electronic devices. The multiresolution image processing technique is based on inherent spatial and spectral distribution properties of the retinal neurons and henceforth aids to achieve a near natural signal feed for Retinal Prosthesis.

nerve to the brain requires that numerous sensory neurons are stimulated in parallel and spatially correct order to enable an accurate encoding of 3-D objects. In visually impaired patients with AMD and RP, the photoreceptors cease to function, although the retina remains intact. There have been several explorations over the past decade by stimulation of the photoreceptors by implanting electrodes inside the retina to generate neural signals for finding out methods of restoring vision in patients suffering with AMD and RP. Broadly, there are two hardware configurations proposed for a visual prosthesis. The system can either be a totally intraocular or can combine intraocular and extraocular hardware [3,4,5]. The latter approach incorporates an extraocular camera and an image processing unit in addition to the retinal stimulation device. This topology is convenient as it increases the image processing capability while reducing the amount of hardware contained within the eye [6]. In context, an image generated by a CCD camera could have a resolution anywhere from 336×224 to 3060×2040 pixels while the electrode array may be limited to only 25×25 electrodes. The image processing demands the spatial reduction in resolution from at least 336×244 to 25×25 while maintaining the effective visual information [7]. The activation levels of the electrodes being as low as four, the method also requires a reduction in the spectral resolution of the image from 256 gray levels to four. Such a reduction is achieved by progressively selecting pixels in the image between an upper and lower brightness threshold and setting them to a single gray level [8].

Keywords— Biomedical image processing, Image analysis, Image coding, Image resolution, Image sampling, Prosthetics, User interfaces.

I.

INTRODUCTION

V

ISION is an enormously complex form of information processing that depends on the retina, a remarkable neuroprocessor located at the rear part of the eye. Perception of sight is initiated when light passing through the pupil of the eye is focused by the lens on the retina’s sensory neuroepithelium. The result of this projection is an inverted image incident on about 130 million photorepectors. These compressed electrical signals are then carried on to the visual cortex of the brain for the perception of vision [1]. Blindness can result when any step of the optical pathway—the optics, the retina, the optic nerve, visual cortex, or other cortical areas involved in the processing of vision—sustains damage. Blinding diseases, such as Retinitis Pigmentosa (RP) and Age-related Macular Degeneration (AMD), cause progressive degeneration of the outer retina. Although there are many examples of electrical devices that can support or replace the function of defective tissues— such as cochlear implants for the hearing impaired [2], or pacemakers for individuals with heart disease—restoring vision with electrical devices implanted into the retina is a comparably difficult task. The transformation of visual signals into appropriate electrical signals to be carried by the optic

II.

EVOLUTION OF THE CONCEPT

Tassicker’s patent in 1956 described how a small, flat, light-sensitive selenium cell placed behind the retina of a blind patient transiently restored the patient’s ability to perceive the sensation of light [9]. Later attempts to restore vision by coupling electrodes to the surface of the visual cortex of blind patients did not provide useful images because of the limited spatial resolution and the fading of phosphenes [10]. Hence, explorations for restoring an effective perception of vision needs a serious consideration of the science of Human Vision. The following article explores the science of Human Vision and inherently mimics the natural technique for restoring the perception of vision in patients suffering from AMD and RP. 1

Multi-Resolution Image Processing in selecting the Most Effective Visual Information for Retinal Prosthesis

Science of Human Vision

Essay

growth of regions have been suggested by Gilmont et al. as early as 1996 [7]. The present work is based on the suggestions made by Gilmont et al., as well as incorporating certain more techniques for decimation of the acquired image in spatial as well as spectral domain.

Vision is a perception different from other physical perceptions like touch, smell and sound in the way, that it involves a complex assembly of optically sensitive neurosensory layer and an associated electrical transmission mechanism. Seeing is initiated when light passing through the dilated disc of the pupil focuses on the retina after passing through the lens. The result of this projection is a reduced up-side down image of the object onto the roughly 130 million photoreceptor cells (rods and cones) in the outermost layer of the retina. The cones are responsible for providing chromatic images of high spatial resolution, and the rods are required for achromatic vision with less spatial resolution in dim light, transform local luminance and color patterns of the projected image into electrical and chemical signals [11]. The cones in each eye number between 6–7 million and are primarily located at the central portion of the retina, called the fovea. Humans can resolve large details through the cones as each one is independently connected to its own nerve ending. The number of rods is much larger between 75–170 million and widely distributed over the retinal surface. The large area of distribution and the fact that several rods are connected to a single nerve end reduce the amount of details discernable by the photoreceptors [12]. These signals then activate a complex circuit of retinal neurons: horizontal cells, bipolar cells, amacrine cells, and ganglion cells. Visual information from the retina’s 130 million photoreceptors is compressed into electrical signals carried by 1.2 million highly specialized ganglion neurons, whose axon for the optic nerve. The optic nerve transmits visual information via the lateral geniculate nucleus to the primary visual cortex of the brain.

IV.

RESULTS AND DISCUSSIONS

Fig.1. Input image to the system

Retinal Prosthetic Implants The feasibility of an implantable retinal prosthesis, which promises to partially restore vision by direct electrical stimulation of the retinal neurons, is supported by several studies. Morphometric analysis in post-mortem eyes with almost complete photoreceptors loss either due to AMD and RP has shown as many as 90% of the inner retinal neurons remain histologically intact [13]. Several research groups have investigated various aspects of retinal prostheses, ranging from electrical stimulation of retinal neurons to surgical implantation methods [3,14,15]. Over the past decade, there have been significant developments in the field of retinal prosthetic methods. The primary approaches are can be classified primarily as extraocular or intraocular. III.

Fig.2. Output of the system

Although the static resolution of the final image is poor (Fig.2), the ability to centre an object in the multi-resolution grid greatly improves the visual quality. In this way, it is possible to scan the desired parts of the object to help any identification, by using the better resolution of the central window. Simulations on the image sequences have shown that the motion of successive images enhances the perceived resolution due to the temporal smoothening operation performed by the brain.

PROCESS OVERVIEW

The primary aim of the work is concerning the signal processing unit which is capable of selecting the most effective visual information for feeding thought to the retinal prosthetic unit. Early explorations based on

2

Essay

Debdoot Sheet and Manjunatha M.

V.

CONCLUSION

The system presented has the advantage of being incorporated on a low-power, small-size chip, and producing images which are useful to blind people despite their low resolution. This process has been executed in real-time and adapts its parameters to the inputs, so that the quality is preserved in environments of variable visibility and lighting conditions.

[8].

[9]. [10].

[11].

REFERENCES [1].

[2].

[3].

[4].

[5].

[6].

[7].

Zrenner, Eberhart. "Will Retinal Implants Restore Vision?" Science, February 8, 2002, Science Bodybuilding: The Bionic Human ed.: 1022-1025. Rauschecker, J P, and R V Shannon. "Sending Sound to the Brain." Science Bodybuilding: The Bionic Human (February 2002): 1025-1027. Mahadevappa, Manjunatha, James D Weiland, Douglas Yanai, Ione Fine, Robert J Greenberg, and Mark S Humayun. "Perpetual Thresholds and Electrode Impedance in Three Retinal Prosthesis Subjects." IEEE Trans. Neural Systems and Rehabilitation Engineering 13, no. 2 (June 2005): 201-206. Weiland, James, and Mark S Humayun. "Artificial Vision by Electrical Stimulation of the Retina." Int. Conf. Neural Nets. Montreal, Canada: IEEE, 2005. 3100-3102. Stieglitz, Thomas, and Joerg Uwe Meyer. "Biomedical Microdevices for Neural Implants." In BioMEMS, edited by G Urban, 71-137. Netherlands: Springer, 2006. Yoon, I Yong, et al. "A new image signal transfer method using laser in artificial retina." 1st Int. IEEE EMBS Conf. Neural Engg. IEEE, 2003. 208-210. Gilmont, T, X Verians, J D Legat, and Cl Veraart. "Resolution Reduction by Growth of Zones for Visual

[12].

[13].

[14]. [15].

Prosthesis." Int. Conf. Image Processing. Lausanne: IEEE, 1996. 299-302. Naghdy, Golshah. Selecting the most effective visual infromation for retinal prosthesis. 2008. http://spie.org/x8568.xml?highlight=x2408 (accessed September 5, 2008). Tassicker, Graham Edward. Retinal Stimulator. United States of America Patent 2,760,483. October 20, 1954. Hallum, L E, S L Cloherty, and N H Lovell. "Image Analysis for Microelectronic Retinal Prosthesis." IEEE Trans. Biomedical Engg. 55, no. 1 (January 2008): 344346. Prives, M, N Lysenkov, and V Bushkovich. The Organ of Vision. Vol. 2, in Human Anatomy, by M Prives, N Lysenkov and V Bushkovich, 399-417. Moscow: MIR Publishers, 1989. Gonzalez, Rafael C, and Richard E Woods. "Digital Image Fundamentals." Chap. 2 in Digital Image Processing, by Rafael C Gonzalez and Richard E Woods, 35-103. Delhi, Delhi NCR: Pearson Education, 2008. Kim, S Y, S Sadda, J Pearlman, Mark S Humayun, E de Juan, and B M Melia. "Morphometric analysis of the macula in eyes with disciform age-related macular degeneration." Retina 22, no. 4 (2002): 471-477. Humayun, Mark S. "Intraocular retinal prosthesis." Trans. American Opthalmological Society 99 (2001): 271-300. Humayun, Mark S, et al. "Visual perception in a blind subject with a chronic microelectronic retinal prosthesis." Vision Research (Elsevier) 43 (2003): 2573-2581.

Corresponding Author: Author: Debdoot Sheet Institute: School of Medical Science and Technology, Indian Institute of Technology, Kharagpur. City: Kharagpur. Country: India e-mail: [email protected]

3