HCC: Medium: Removing Barriers to the Practical Use of Non-Invasive Brain-Computer Interfaces

点击翻译按钮获取中文摘要

Brain-computer interfaces (BCIs) are hardware and software systems that allow users to interact with computer applications by changing their mental activity, which causes variations in weak electrical voltages produced by the brain. BCIs measure these voltages in one of two ways: invasive methods use electrodes implanted in the brain, while noninvasive methods use electrodes resting on the scalp that are part of a cap worn by the user. A long-term goal of BCI research is a new mode of communication for subjects with diseases and injuries resulting in the loss of voluntary muscle control, such as amyotrophic lateral sclerosis (ALS), multiple sclerosis, high-level spinal cord injuries or severe cerebral palsy. If all voluntary muscle control is lost, a locked-in syndrome results in which a person is unable to communicate with the outside world. BCIs can provide a new way for users to communicate with their caregivers and to control devices such as televisions, wheelchairs, speech synthesizers and computers. While BCI technology holds great promise, most BCI systems remain in research labs. The goal of this project is to remove barriers to practical, noninvasive, BCI technology that exist in current approaches, and to field test the resulting BCI systems in the homes of users who suffer from motor impairments. Limitations of current BCI systems that will be addressed include the difficulty of applying an electrode cap, signal artifacts due to other assistive technology in the user's environment, and long computer and user training times required to calibrate current EEG classification algorithms.A key barrier to practical BCI systems is the lack of methods for reliable, fast classification of EEG signals. In this project, this limitation will be addressed by conducting experiments in three areas. One set of experiments will investigate the quality of EEG signals recorded in subjects' homes and the performance of BCI applications in real-time in the homes. The second set of experiments will involve new algorithms for EEG artifact removal and signal classification that are tailored for EEG recorded in subjects' homes and for real-time use. For the second set of experiments, new user interfaces will be studied and compared to currently available interfaces. For the third set of experiments, several different user interface designs for BCI applications will be developed and studied. The effectiveness of visual and auditory feedback provided to the user in real-time will be investigated. This interdisciplinary project involves a team of investigators and students from diverse backgrounds. Faculty and students in computer science will design and implement algorithms and the BCI user interface. Faculty and students in occupational therapy will guide the field testing of BCI systems and will guide the evaluation of these experiments. Progress will be evaluated in a number of ways, including experiments comparing EEG signal representations and classifiers by accuracy, reliability, and training time, and field tests of BCI systems. Ultimately, the project's success will be measured by new or improved means of individuals interacting with computers in their homes for purposes of communication with others and control of assistive devices like wheelchairs.Broader Impacts: This project will develop a new technology for sensing and analyzing electroencephalogram signals (EEG) from human subjects. The resulting technology will help advance brain imaging and its application. The long term goal of this research is a new brain-computer interface based on EEG signals with which persons can use a computer to communicate with others in their vicinity or remotely over the net, to surf the net, and to control environmental entertainment, and assistive devices. The new technology will be simple enough for any person with minimal training to use. The project will also play a strong role in the education of future researchers and health professionals in this interdisciplinary field by involving graduate and undergraduate students from multiple departments as research assistants, by teaching a new course in BCI for students from a variety of backgrounds, and by providing fieldwork experiences.