Bionic eye technology that could restore vision to millions is the goal of the Monash vision group, a research team whose expertise spans engineering, neurosurgery and physiology. Its creation, a device that stimulates the brain rather than the retina, is on track for trial in just two years.
Ever since Steve Austin ran across our screens in the 1970s television series The Six Million Dollar Man, the notion of a 'bionic eye' that allows the blind to see has tantalised our imaginations. But how to achieve it? The fictional Austin had a high-powered camera implanted in his eye, but in real life it is not so simple. Replicating the process by which we make sense of the visual world involves a complex interface between the brain and a machine.
More than a dozen bionic eye projects are at various stages around the world, each attempting to use a prosthesis to replicate the mechanics of 'seeing', generally by stimulating the nerve cells in the retina.
But the interdisciplinary Monash Vision Group (MVG) is bypassing the eye altogether by using a device that aims to stimulate the visual cortex area of the brain. This powerful implant will communicate directly with the brain to re-create a form of vision and could restore sight to 85 per cent of the millions of people who are clinically blind.
The device designed by MVG consists of a digital camera mounted into a pair of spectacles that sends visual information to a pocket-sized processor. It converts the image into signals that are transmitted wirelessly to an implant at the back of the brain. A prototype device is likely to be ready for implantation into a human volunteer by 2014.
The MVG was established in April 2010 with an A$8 million grant from the Australian Research Council. It is directed by Professor Arthur Lowery of the Department of Electrical and Computer Systems Engineering and managed by Dr Jeanette Pritchard. The MVG is an extensive team effort that involves more than 20 people from Monash University and The Alfred hospital across disciplines including physiology, neurosurgery, electronic engineering, materials engineering and immunology. Private companies Grey Innovation and MiniFAB are also part of the core team – designing, manufacturing and commercialising the device.
As a systems engineer, Professor Lowery says his role has been to integrate knowledge from myriad areas. His background is in optical equipment systems, but he has had to acquire a working knowledge in areas as diverse as brain electrophysiology, robotic vision processing, materials technology and medical device technology. "The systems engineer is at the top of the design hierarchy. My job is to make sure every design decision takes account of the many limitations we are working under."
Most of the elements involved in the process are not new, Professor Lowery says. "Individually, each problem involves basic principles that we already know about. It was about bringing all these together to create something that had never been done before."
Among the experts who have contributed to the project is Professor Marcello Rosa, from the Department of Physiology at Monash. Professor Rosa has spent much of his career investigating and mapping the parts of the brain that make sense of vision, and says he was brought in to answer a question: what is the most promising part of the brain for the best result with the least cost and complexity of hardware?
The area of the retina that gives high-acuity vision is very small, limiting the ability of retinal implants to produce detailed images, Professor Rosa says. Moreover, implants can only work if the retinas are still healthy. The visual cortex, by contrast, is conveniently located at the back of the brain and offers more than 20 times the available surface area for high-acuity vision.
Experiments in the 1980s showed it was possible to generate flashes of light in the brain by stimulating a point in the visual cortex. Professor Rosa has built on that knowledge to develop a device to stimulate not just one speck of light, but hundreds. These will work in the user's brain like pixels within a grid that can be used to 'draw' images – a little like the LED traffic signs that convey information to road users.
"Given that each speck of light is stimulated by an electrode, the more electrodes that can be implanted in the brain the more detailed the picture that can be drawn," Professor Rosa says. Working within the limitations of miniaturisation and the physical stability of the implant, the team has come up with a device that will deliver 600 electrodes to the brain via 15 tiny tiles, each the size of the tip of the little finger.
Generating images that will be meaningful within that grid of 600 dots of light is the job of Dr Wai Ho Li, a lecturer and researcher in the Department of Electrical and Computer Systems Engineering at Monash. Dr Li's background is in intelligent robotics and visual perception. "My role is to translate the image from a camera into something the brain can recognise and understand," he says.
Early bionic eye systems reproduced the visual world as light and shade, but the lack of detail in such images made them impractical for the user. To provide a more meaningful image, Dr Li came up with the concept of 'transformative reality', which turns the visual world into a symbolic language.
"Our aim is to help people do things they can't do with a white cane and a seeing-eye dog," he says. "Things like being able to navigate through cluttered indoor spaces, for instance, or find their coffee cup on their desk, or notice if someone is waving."
Depending on which mode it is in, the processor 'looks' at a scene, strips out the clutter and communicates the pertinent information to the user. In 'object identification mode' the focus is on the edges of three-dimensional objects – a coffee cup on a table, for instance. In 'empty ground mode', a clear path through a corridor is illuminated rather like the way an exit-path is lit in an aeroplane. 'Person identification mode' can convey a crude image of a face and body and simple gestures – although not as yet facial recognition or expressions.
Dr Li's involvement with the project extends from the camera through to the processor. Designing the implant circuitry itself and ensuring it is strong and reliable enough to be long-lasting, as well as small enough to implant in the brain, is the job of Dr Jean-Michel Redouté, a senior lecturer in the Department of Electrical and Computer Systems Engineering at Monash. He says his team's challenge has been to ensure the implant can carry out highly complex tasks within an extremely compact package that is biologically inert and, most importantly, does not demand too much power.
"Because the implant is designed to work wirelessly, it needs to be powered wirelessly as well," he says. "So our effort has been to design the circuitry in such a way that it allows a high transmission of data but keeps power consumption low."
One of the important parts of the puzzle was determining what it is feasible to implant in a human brain. Professor Jeffrey Rosenfeld, head of the Department of Surgery at Monash and director of neurosurgery at The Alfred hospital, has been involved since the beginning of the project to ensure non-medical engineers were aware of the biological limitations. "It's all very well to design a beautiful electronic gadget, but unless it's applicable to the human brain it is no use to anyone."
Apart from advising on the materials suitable for implantation into the brain, Professor Rosenfeld is ensuring other surgeons understand and can follow the process. "It's no good if I am the only person who is able to insert the device," he says. Among the many sub-projects is the development of an insertion tool that will deliver the implant into the brain with the exact force required to penetrate the electrodes to a specific depth without causing trauma.
Professor Rosenfeld says the project stands on the shoulders of Professor Graeme Clark, the inventor of the Cochlear implant – also known as the bionic ear. "We are not starting from scratch here, in that he pioneered the notion of implanting electrodes to stimulate specific points in the brain." However, the bionic eye project is far more complex. The Cochlear implant requires 23 electrodes to convey signals to the brain; 600 electrodes are needed for vision.
After two years of research, the next step is to ready the device for implantation into a human brain. What its users will actually 'see' is the ultimate test, but Professor Lowery believes there is every reason to be optimistic. "We know the fundamental principles work and we're confident this device is going to give people who are blind a lot more independence."