Working in interdisciplinary teams from seven European universities and institutions with complementary expertise in computational, systems and clinical neuroscience, materials engineering, microsystems design and deep learning, the project will develop technology to restore vision of blind people through electrical stimulation of the brain.
The aim of the project is to develop a neuroprosthesis with thousands of electrodes driven by adaptive machine learning algorithms for a new brain-computer interfacing technology. “We want to create a novel neuroprosthesis system that is lightweight, robust and portable, and which will remain effective for decades,” explains Shih-Chii Liu. Current systems only stimulate a small set of neurons in the brain, and interfaces have a longevity of only a few months.
Liu is convinced that the project will succeed in its goals: “All the partners have long-time experience in their respective fields, so the required background knowledge is already in place. The breakthroughs will come with the planned larger scale efforts and partner interactions in this project.” The challenge will be coordinating the expected breakthroughs across multiple disciplines.
These breakthroughs include innovative approaches for stimulation with high-electrode-count interfacing with the visual cortex. For this, thin flexible probes are needed that cause minimal tissue damage as well as new electrode coatings and novel microchip methods. The researchers will also channel the stimulation currents to many thousands of electrodes and monitor neuronal activity in higher cortical areas.
Breakthroughs are also expected when it comes to artificial neural networks trained by deep learning, which will only extract the most relevant visual information from a camera input to enable blind individuals to recognize objects and facial expressions and navigate through unfamiliar environments. These networks will transform the camera footage into stimulation patterns that drive the neurons in a way that the blind person can interpret. This is the only way that the signals can be processed and passed on. At the same time, eye tracking will be used to improve perception in a closed-loop approach.
MEDICA-tradefair.com; Source: University of Zurich
