Skip to content. Skip to navigation
CIM Menus
 

Neural Interfaces: From Robot Eye to Human Eye


Dr. Tohru Yagi
The Institute of Physical and Chemical Research
Japan

January 31, 2005 at  2:00 PM
MD 357

My main research interests are biological and artificial vision systems. The goal of my research is to understand the neural mechanisms and computational principles of biological vision and to exploit its findings in engineering applications especially with relation to biomedical and rehabilitation engineering and bio-mimetic robotics. My approach involves a combination of engineering and biomedical approaches, based on vision science expertise. I have been engaged in the following research topics throughout my career in academia and industry; a visual prosthesis, a human interface using biological signals, fixation study, and a bio-mimetic robot vision system. In my talk, I will present the first two topics mainly and possibly show some demonstrations.


1) Visual Prosthesis

A visual prosthesis is a microelectromechanical system (MEMS) to restore vision in blind patients by applying electrical stimulation to the visual nervous system. In academia, I have conducted basic research and system design and integration on a bio-hybrid retinal implant, which consists of cultured neurons on MEMS. In this implant, axons of the cultured neurons are guided toward the central nervous system (CNS) using a peripheral nerve graft. Because the cultured neurons make functional connections between the MEMS and the CNS, electrical stimulation causes the cultured neurons to send visual information to the CNS in the bio-hybrid retinal implant.


2) Bio-signal Human Interface

I have developed an eye-gaze interface using EOG (electro-oculography), a biological signal related to eye movements. Using the EOG switch a user can easily turn on/off a nurse-call device using just eye movements. This was commercialized by a start-up company, and is now widely used among ALS (amyotrophic lateral sclerosis) patients in Japan. I have also developed highly-functioned eye-gaze interface. This device enables a user to move a computer cursor using only eye-gaze. In order to overcome drifting and blinking problems, a unique method of calibration and signal processing was developed. In addition to the EOG interface, I have conducted basic research on EEG (electro-encepharography) related to eye movements. I have found that EEG potential at the occipital lobe changes rapidly just before an eye movement starts, and the change is observed at the right occipital lobe when the eyes move toward the right direction, and at the left, toward the left.