echemi logo
Product
  • Product
  • Supplier
  • Inquiry
    Home > Active Ingredient News > Study of Nervous System > Nature: The brain-computer interface can convert brain activity into text in real time!

    Nature: The brain-computer interface can convert brain activity into text in real time!

    • Last Update: 2021-05-21
    • Source: Internet
    • Author: User
    Search more information of high quality chemicals, good prices and reliable suppliers, visit www.echemi.com

    Scientists are exploring ways to allow people with disabilities to exchange ideas with them.


    PIXABAY

    PIXABAY

    The brain-computer interface (BCI) enables people who have lost the ability to move or speak to resume communication.


    Recently, researchers from Stanford University in the United States combined artificial intelligence (AI) software with brain-computer interface (BCI) equipment for the first time, and successfully developed an intracortical BCI that decodes neural activity in the motor cortex to try handwritten actions and uses cyclic nerves.


    Successfully developed an intracortical BCI, which attempts handwritten actions by decoding neural activity in the motor cortex, and uses a recurrent neural network decoding method to convert it into text in real time.


    The research team worked with a participant in a clinical trial named BrainGate2.


    Researchers found that because the peaks and valleys of neural activity vary with time, perhaps due to fluctuations in writing speed, neural activity seems to be strong and repeatable.


    In order to determine whether the neural activity encodes the pen tip movement required to draw each shape, the researchers reconstructed each character by linearly decoding the pen tip speed from the experimental average neural activity.


    Neural representation of handwriting

    Neural representation of handwriting

    Second, the researchers used the non-linear dimensionality reduction method (t-SNE) to visualize the neural activity of each experiment in two dimensions (2D), and record the relevant information after giving the subject a "go" prompt.


    The t-SNE method shows the tight clusters of neural activity of each character and a dominant motion encoding.


    As a result, the researchers concluded that even after years of paralysis, the neural representation of handwriting in the motor cortex may still be powerful enough to be expressed through brain-computer interface technology.


    Even after many years of paralysis, the neural representation of handwriting in the motor cortex may still be powerful enough to be expressed through brain-computer interface technology.


    Krishna Shenoy, a researcher at the Howard Hughes Medical Institute at Stanford University, co-author of the research report, said that with further development, this innovation will allow paralyzed people to type quickly without using their hands.


    The next step for the team is to work with participants who cannot speak, such as those with amyotrophic lateral sclerosis, a degenerative neurological disease that causes loss of movement and language.


    Jose Carmena, a neuroengineer at the University of California, Berkeley, said, "This technology and other similar technologies have the potential to help people with various disabilities.


    Original source:

    Francis R.


    Francis R.
    Willett, Donald T.
    Avansino, Leigh R.
    Hochberg, Jaimie M.
    Henderson, Krishna V.
    Shenoy.
      High-performance brain-to-text communication via handwriting .
      Nature , 2021; 593 (7858): 249 DOI:  doi.
    org/10.
    1038/s41586-021-03506-2">10.
    1038 / s41586-021-03506-2 High-Performance Brain-to-text Communication Via Handwriting Nature doi.
    org/10.
    1038/s41586-021-03506-2">10.
    1038 / s41586-021-03506-2 10.
    1038 / s41586-021-03506-2

    in this message
    This article is an English version of an article which is originally in the Chinese language on echemi.com and is provided for information purposes only. This website makes no representation or warranty of any kind, either expressed or implied, as to the accuracy, completeness ownership or reliability of the article or any translations thereof. If you have any concerns or complaints relating to the article, please send an email, providing a detailed description of the concern or complaint, to service@echemi.com. A staff member will contact you within 5 working days. Once verified, infringing content will be removed immediately.

    Contact Us

    The source of this page with content of products and services is from Internet, which doesn't represent ECHEMI's opinion. If you have any queries, please write to service@echemi.com. It will be replied within 5 days.

    Moreover, if you find any instances of plagiarism from the page, please send email to service@echemi.com with relevant evidence.