Brain Computer Interface - BCI

---Compile Silicon Valley live
  brain-computer interface (English: brain-computer interface, referred to as BCI; sometimes called direct neural interface or brain-machine interface), is a human or animal brain (or culture of brain cells) and external A direct connection path created between devices. In the case of a one-way brain-computer interface, the computer either accepts commands from the brain or sends signals to the brain (such as video reconstruction), but cannot send and receive signals at the same time. A two-way brain-computer interface allows two-way information exchange between the brain and external devices.
In this definition, the word "brain" means the brain or nervous system of an organic life form, not just an abstract "mind". "Machine" means any processing or computing device that can take the form of a simple circuit to a silicon chip.
Research on brain-computer interfaces has been going on for more than 30 years. Such knowledge gained from experiments has grown significantly since the mid-1990s. Based on years of practice in animal experiments, early implantable devices for humans were designed and manufactured to restore impaired hearing, vision, and limb motor abilities. The main line of research is the unusual cortical plasticity of the brain, which is adapted to the brain-computer interface and can control the implanted prosthesis like a natural limb. With the current advances in technology and knowledge, the pioneers of brain-computer interface research can convincingly try to create brain-computer interfaces that enhance human function, not just restore human function. This technology has previously only existed in science fiction.


Brain-computer interface and neural repair
Neural repair is a field related to nerve repair in neuroscience, that is, replacing part of nerves or sensory organs whose original functions have been weakened with artificial devices (prostheses). The most widely used neuroprosthesis is the cochlear implant, which has been implanted in about 100,000 people worldwide by 2006. There are also neuroprostheses used to restore vision, such as artificial retinas, but current work in this area is limited to implanting artificial devices directly into the brain.
The difference between brain-computer interfaces and neuroprosthetics is mostly literal: "Neuroprosthetics" generally refers to devices used clinically, while many existing brain-computer interfaces are still experimental. In practice, neural prostheses can be connected to any part of the nervous system, such as the peripheral nervous system; while "brain-computer interface" usually refers to a narrower class of systems that are directly connected to the brain.
The terms "neural repair" and "brain-computer interface" are often used interchangeably because of similarities in goals and means of achieving them. Neuroprosthetics and brain-computer interfaces attempt to achieve a common goal, such as restoring the ability to see, hear, motor, and even cognition. Both use similar experimental methods and surgical techniques.
Animal brain-computer interface research[edit]
Some laboratories have achieved recording signals from the cerebral cortex of monkeys and rats in order to operate brain-computer interfaces for motor control. The experiment had the monkeys manipulate an on-screen computer cursor and control a robotic arm to perform simple tasks simply by recalling a given task (without any movement happening). Additional research on cats decoded visual signals.


Motor function-oriented brain-computer interface
In terms of motor function-oriented brain-computer interface, the development of algorithms to reconstruct the control of motor cortex neurons on movement can be traced back to the 1970s. A group led by Schmidt, Fetz and Baker demonstrated in the 1970s that monkeys can rapidly learn to freely control the firing frequency of individual neurons in the primary motor cortex following closed-loop operant conditioning. In the 1980s, Apostolos Georgopuolos of Johns Hopkins University found a relationship between the direction of upper limb movements in macaque monkeys and the firing patterns of individual neurons in the motor cortex. He also found that a discrete set of neurons could also encode limb movements.
Since the mid-1990s, motion-oriented brain-computer interfaces have experienced rapid development. Several research groups have been able to use neural swarm recording techniques to capture complex neural signals in the motor cortex in real time and use them to control external devices. Among them are the research groups of Richard Andersen, John Donoghue, Phillip Kennedy, Miguel Nicolelis and Andrew Schwartz.

Brain-computer interface for sensory functions
Currently humans have been able to repair or are trying to repair include auditory, visual and vestibular sensations.
Cochlear implants are by far the most successful and clinically used brain-computer interface.
Vision restoration technology is still under development. The main reason why the research and application of this aspect lag behind the auditory equivalent is the huge amount of information transmitted by vision and the relative complexity of the functions of the peripheral sensory organs (retina) and the central visual system. See Vision Prosthetics for details.
Della Santina and colleagues at Johns Hopkins University have recently developed a vestibular implant that restores three-dimensional vestibular sensation.

Research Progress Milestones
Phillip Kennedy and colleagues used neurotrophic-cone electrode implantation to build the first intracortical brain-computer interface in monkeys.
In 1999, Garrett Stanley of Harvard University attempted to reconstruct visual images by decoding the firing of neurons in the lateral geniculate body of the cat's thalamus. They recorded the spikes of 177 neurons and used filtering to reconstruct eight videos played to the cat, from which recognizable objects and scenes could be seen.
Miguel Nicolelis of Duke University is among the representatives who support the use of electrodes covering vast cortical areas to extract neural signals and drive brain-computer interfaces. The advantage of this approach, he believes, is the ability to reduce the instability and randomness of neural signals captured by a single electrode or a small number of electrodes. After initial studies in rats in the 1990s, Nicolelis implemented experiments in night monkeys that could extract signals from cortical motor neurons to control a robotic arm. By 2000, Nicolelis' group had successfully implemented a brain-computer interface that reproduced arm movements in night monkeys as they manipulated a joystick to get food. This brain-computer interface can work in real time. It can also remotely control the robotic arm via the Internet. However, since the monkeys themselves do not receive sensory feedback from the robotic arm, this type of brain-computer interface is open-loop. Later work by Nicolelis' group used rhesus monkeys.
Other labs designing brain-computer interface algorithms and systems to decode neuronal signals include John Donoghue of Brown University, Andrew Schwartz of the University of Pittsburgh, and Richard Anderson of Caltech. The researchers' brain-computer connections used 15-30 neurons at a time, significantly less than Nicolelis's 50-200. The main work of Donoghue's group is to achieve the rhesus monkey's motor control of the cursor on the computer screen to track the visual target. The monkeys did not need to move their limbs. [11] The main work of the Schwartz group is the visual target tracking in the three-dimensional space of virtual reality, and the control of the robotic arm by the brain-to-brain interface. [12]. This group claims that their monkeys can feed themselves zucchini through a robotic arm controlled by a brain-computer interface. [13] Anderson's group is working on a brain-computer interface that extracts premotor signals from neurons in the posterior parietal lobe. Such signals include those produced by experimental animals in anticipation of a reward.
In addition to the above-mentioned brain-computer interfaces for calculating motion parameters of limbs, there are also brain-computer interfaces for calculating electrical signals (electromyography) of muscles. One promising application of this type of brain-computer interface is to stimulate the muscles of paralyzed patients to recreate the function of their voluntary movements.

Human Brain-Computer Interface Research
Invasive Brain-Computer Interface
Invasive brain-computer interfaces are mainly used to reconstruct special senses (such as vision) and motor functions in paralyzed patients. Such brain-computer interfaces are usually implanted directly into the gray matter of the brain, so the quality of the acquired neural signals is relatively high. But its disadvantage is that it is easy to trigger immune response and callus (scar), which leads to the decline or even disappearance of signal quality.
A pioneer in visual brain-computer interfaces was William Dobelle. His cortical visual brain-computer interface is mainly used in patients with acquired blindness. In 1978, Dobelle implanted an array of 68 electrodes in the visual cortex of a blind male Jerry, and successfully created Phosphene. The brain-computer interface system includes a video camera, signal processing device and driven cortical stimulation electrodes. After implantation, the patient can see grayscale-modulated, low-resolution, low-refresh-rate bitmap images within a limited field of view. The visual prosthesis system is portable and can be used independently by the patient without the assistance of physicians and technicians.
In 2002, Jens Naumann became the first of 16 patients to be implanted with Dobelle's second-generation cortical vision prosthesis. The second-generation cortical vision prosthesis features better mapping of phosphenes to the visual field, creating more stable and uniform vision. Its optical illusion lattice covers a larger field of view. Soon after receiving the implant, Jens was able to roam around the research center at a slow speed by herself.
For the brain-computer interface of "motor nerve prosthesis", Philip Kennedy and Roy Bakay of Emory University are the first to implant an invasive brain-computer interface in humans that can obtain neural signals of sufficient quality to simulate movement. Their patient, Johnny Ray, suffers from locked-in syndrome caused by a stroke in the brainstem. Ray received the implant in 1998 and survived long enough to learn to use the brain-computer interface to control a computer cursor.
In 2005, Cyberkinetics was approved by the US FDA to conduct the first phase clinical trial of the motor cortex brain-computer interface in nine patients. Matt Nagle, a quadriplegic, became the first patient to use an invasive brain-computer interface to control a robotic arm. He was able to complete tasks such as robotic arm control and computer cursor control through motor intent. Its implants are located in the areas of the motor cortex of the anterior middle gyrus that correspond to the arms and hands. The implant, called BrainGate, is an array of 96 electrodes.
The above information is excerpted from Wikipedia.

Silicon Valley Live
is a paid-for knowledge product of Silicon Valley spy. It is positioned as a global innovation practice sharing, and delivers Silicon Valley's cutting-edge technology, industrial innovation, and innovative thinking to the global Chinese. On June 28, 2017, Dr. Han Bicheng, a Harvard brain-computer interface scientist, will launch the "Brain-Computer Interface Series Course". You will receive a world-first brain-computer interface industry map and other heavy benefits.

Silicon Valley Detective,
founded in September 2015 in San Mateo, California, is a new media focusing on global innovation and Silicon Valley technology venture capital. Established one and a half years ago, the entire network has more than 3.5 million subscribers. Won the 2016 Tencent T+ New Media Award and the 2016 New List Annual Technology New Media Honor Award.