Strategies for the development of a color visual prosthesis for the blind


Advancements in the medical field have allowed scientists to develop artificial devices to replace various human body parts or enhance the functionality of the existing organs and tissues. Although prostheses of some human body parts have undergone tremendous development, others are yet to be fully developed. In particular, the idea of a human color visual prosthesis has recently drawn research attention. So far, all implanted human prosthetic visual systems have only provided achromatic artificial visual experiences with relatively low spatial resolution. With the continued search for more effective and efficient visual prosthetic systems, devices eliciting colored phosphenes will be of interest in the future.1

The elicitation of white phosphenes has been extensively studied in the past. However, colored phosphenes can be elicited by focal electrical stimulation of human visual cortex (Figure 1). Although about 52% of phosphenes are a single color, about 75% exhibit phosphenes with some color features.2 Previous studies have employed artificial visual systems utilizing relatively large surface electrodes and high stimulation intensities. However, using penetrating electrodes, phosphenes exhibit a variety of colors that desaturate to white when the stimulus intensity is increased.3 This suggests unnecessary over-stimulation of the visual cortex by the previously reported B/W prosthetic systems designed to achieve white phosphenes. Although this hypothesis is yet to be conclusively demonstrated, it has been speculated to be the reason why seizures and alterations in phosphenes are not uncommon occurrences.

Generally, the differences observed in the phosphene colors can be attributed to two main factors: unnecessarily high stimulus intensities and the unnecessarily large size of the activated cortical area being stimulation. It appears more advantageous for a color prosthesis to be employed using low stimulation levels, reducing the likelihood of epileptogenic responses. To enhance the development of a color visual prosthesis, Professor Vernon Towle and his colleagues from the University of Chicago and Illinois Institute of Technology have proposed a new hybrid simulation model. The approach combines B/W and color stimulation to provide color information without a loss of spatial resolution. A uniform random distribution in the red-blue-green (RBG) space was adopted to model color selectivity in the cortex. Specifically, the authors discussed the ideas of software implementation approaches and the merits and demerits of improved color phosphene visual prostheses. The original research article is currently published in the journal, Journal of Neural Engineering.

The research team showed that real-world colors are diverse and spread along brightness or chromatic gradients (Figure 2). This indicated that the normal perception of real-world images may not be as veridical as has been assumed. Schmidt’s simulation results confirmed previous reports as they demonstrated the possibility of obtaining saturated colored phosphenes with very low stimulation levels.3 The prosthesis system they propose employs two stimulation selection methods, one a B/W mode at relatively higher intensities and another for color mode at low intensities. The color mode exhibited the advantage of low seizures and after-discharges. The advantages of colored phosphenes include improved outdoor environment navigation and the joy of perceiving colors. The hybrid approach emerges as promising in not exacerbating the low spatial resolution associated with the existing systems due. Specifically, the hybrid strategy is that when a colored camera pixel matches the color associated with a stimulating electrode, the color will be perceived by using low-level stimulation; if the colors do not match, higher-level stimulation will be used to elicit either a white phosphene or no phosphene.

In a nutshell, the new article provided useful insights that would significantly contribute to the development of a color visual prosthesis. The authors focused their arguments and ideas on the capability of different software and hardware implementation strategies to incorporate color in developing future prosthetic visual systems. Although software and cameras do not present significant challenges, analyses revealed the diverse nature of real-world colors and parsing their distribution along brightness or color gradients. Otherwise, the main limitation of switching to all color phosphenes is the possible significant decrease in the spatial image resolution. In a statement to Advances in Engineering, Professor Vernon Towle said that the new strategies will advance the development of hybrid systems for enhanced color visual prostheses. A more advanced algorithm is currently in preparation by the research team.

Strategies for the development of a color visual prosthesis for the blind - Advances in Engineering

Strategies for the development of a color visual prosthesis for the blind - Advances in Engineering
Figure 1. Wireless floating microelectrode arrays (WFMAs) containing 16 stimulating electrodes that penetrate visual cortex.4
Strategies for the development of a color visual prosthesis for the blind - Advances in Engineering
Figure 2. Natural images contain a wide variety of
colors that are often arranged in broad intensity or
color gradients in RGB space.1

About the author

Professor Vernon L. Towle has spent most of professional career studying the functional organization of the human brain, with his main interest being the human visual system. For the last 40 years he provided surgical clinical neurophysiology services at the University of Chicago. In 1993, his group published the first functional MRI studies using a normal clinical scanner, followed in 1994, by a highly cited PNAS article revealing the reorganization of the human somatosensory and motor areas after unilateral perinatal stroke. A major theme of his work has been to relate brain function to brain structure. This started in 1993, when he devised the first published technique to merge the spherical neurophysiologic dipole models of the day with magnetic resonance images, bringing together for the first time, two totally separate areas of imaging research. He was the first to display electrocorticographic findings on 3-D MRI models of cortex. More recently, he described the cortical organization and complex dynamics of high-frequency ECoG patterns elicited during language tasks. For the last twenty years, working with Phillip Troyk‘s group at Illinois Institute of Technology, they are bringing an intracortical visual prosthesis for blind people to human trials. Dr. Towle has recently described a strategy to include color perception in this project.1


1Towle, V., Pham, T., McCaffrey, M., Allen, D., & Troyk, P. (2021). Toward the development of a color visual prosthesisJournal of Neural Engineering, 18(2), 023001.

Go To Journal of Neural Engineering

2Penfield, W., Rasmussen, T. (1950) The Cerebral Cortex in Man. Macmillan, New York.

3Schmidt, E., Bak, M., Hambrecht, F., Kufta C., O’Rourke, D., & Vallabhanath, P. (1996) Feasibility of a visual prosthesis for the blind based on intracortical microstimulation of the visual cortex. Brain, 119, 507-522.

4Troyk, P., Frim, D., Roitberg, B., Towle, V., Takahashi, K., Suh, S., Bak, M., Bredeson, S, Zhe, H. (2016). Implantation and testing of WFMA stimulators in macaque. Conf Proc IEEE Eng Med Biol Soc. Aug; 4499-4502. doi: 10.1109/EMBC.2016.

Check Also

Remaining useful life prediction for degradation with recovery phenomenon based on uncertain process - Advances in Engineering

Remaining useful life prediction for degradation with recovery phenomenon based on uncertain process