Articles tagged with "assistive-technology"
New brain-computer tech lets paralyzed patient talk in real time
A new investigational brain-computer interface (BCI) developed by researchers at the University of California, Davis, has enabled a paralyzed patient with Amyotrophic Lateral Sclerosis (ALS) to communicate in real time using a synthesized version of his own voice. ALS causes loss of muscle control, including speech, making communication difficult or impossible. This BCI system uses surgically implanted microelectrode arrays in the brain’s speech region to capture neural activity, which is then decoded by advanced AI algorithms to produce near-instantaneous audible speech. The technology significantly reduces the delay seen in previous speech neuroprostheses, allowing for more natural, spontaneous conversations with a delay as low as one-fortieth of a second. The system was tested on a 45-year-old participant in the BrainGate2 clinical trial, who was asked to attempt speaking sentences displayed on a screen while his brain activity was recorded. The AI model mapped his neural firing patterns to intended speech sounds, enabling the participant to control
robotbrain-computer-interfaceneuroprostheticsreal-time-voice-synthesismicroelectrode-arraysassistive-technologyparalysis-communicationA Neuralink Rival Just Tested a Brain Implant in a Person
Paradromics, an Austin-based neurotechnology company founded in 2015, has conducted its first human test of Connexus, a brain implant designed to restore speech and communication in people with paralysis caused by spinal cord injury, stroke, or ALS. The device translates neural signals into synthesized speech, text, and cursor control by recording electrical activity from individual neurons via 420 tiny electrodes embedded in the brain tissue. The initial human implantation occurred on May 14 at the University of Michigan during epilepsy surgery, where the device was temporarily inserted into the temporal lobe using a specialized EpiPen-like tool. This procedure allowed researchers to confirm the device’s ability to capture neural signals with high resolution, which is critical for accurately decoding intended speech. Connexus is part of a growing field of brain-computer interface (BCI) technologies, including Elon Musk’s Neuralink and Synchron, which also develop implants to interpret neural signals but differ in electrode design and signal resolution. Unlike other devices that record from groups of neurons, Paradromics’ implant targets individual neurons to achieve higher-quality signals. BCIs do not read private thoughts but decode neural patterns associated with intended movements, such as facial muscle activity involved in speech. Recent studies from Stanford and UC San Francisco have demonstrated the ability to decode intended speech at rates approaching half of normal speaking speed in paralyzed individuals. Paradromics aims to launch a clinical trial by the end of 2023 to implant Connexus long-term in patients with paralysis, advancing toward commercial availability despite the regulatory and technical challenges of fully implantable brain devices.
robotbrain-computer-interfaceneural-implantsmedical-devicesneurotechnologyassistive-technologybiomedical-engineeringMIT engineers create elder assist robot E-BAR to prevent falls at home
roboteldercareassistive-technologyfall-preventionmobility-supportMITE-BARChip não giúp chỉnh sửa video và đăng YouTube bằng suy nghĩ
robotIoTNeuralinkbrain-computer-interfaceassistive-technologyAIALSRobot Talk Episode 110 – Designing ethical robots, with Catherine Menon
robot-ethicsassistive-technologyautonomous-systemsAI-safetyhuman-robot-interactionethical-designpublic-trust-in-AI