Some indicators in the body are measured through the microelectromechanical system sensor, and the microelectromechanical system actuator can directly act on the organ or diseased tissue for more direct treatment, and the system can be wirelessly powered by the microelectromechanical system energy harvester, and multiple units can transmit information through the microelectromechanical system communicator to grasp the health of the patient at any time, which has a huge role in promoting the progress of medical treatment.

Although MEMS medical care is still far from being fully applied, it is more socially tolerant and easier to achieve than implanting electrodes directly into the patient's brain.

Although the robotics laboratory is also cooperating with domestic medical universities in the research and development of brain-computer interfaces, it is extremely difficult to capture various signals sent by brain neurons at this stage.

Yang Jie also read the literature on brain-computer interface in his previous life, don't look at the human brain is just a jelly ball-sized thing, there are about 20 billion active neurons in the cortical layer of the outer layer of the brain, the volume of the entire cortex is about 500,000 cubic millimeters, and there are about 20 billion neuronal cell bodies in this space.

It's an epic challenge just to distinguish the different functions of these neuronal cell bodies, and the cell bodies are only a small part of the structure of neurons, and these cells will stick out many twisted and bifurcated dendrites, and these dendrites go down to the spinal cord and body, and they are like a dense mass of charged spaghetti inside the cortex.

Each neuron has up to 1,000 to 10,000 synaptic connections to other neurons, more than 20 trillion independent neural connections, and these neurons in different parts are intertwined with each other.

The voltage of each neuron is constantly changing, and this change frequency can reach hundreds of times per second. And the synaptic connections of these neurons often change size, disappear, and then reappear.

What's even more terrifying is that there are a lot of other cells mixed in the cortex, including something called glial cells, which have many different variants, each responsible for different functions, such as cleaning chemicals released into synapses, wrapping axons with myelin sheaths, and acting as the immune system of the brain, which is about the same number as neurons.

How to distinguish them and record and analyze the bioelectric signals they emit is one of the most difficult challenges in human history at this stage, which may take decades, maybe hundreds of years, and is an impossible task now.

Engineers don't have endless brains to fiddle with in the lab, and most people don't want their heads to be opened by these scientists to study inside.

Now there are two ways to get signals from neurons in the brain: invasive and non-invasive, and there are several types of invasive, but the methods used are almost the same-

These neuroscientists make a metal wire between 10 and 30 microns in diameter out of gold, platinum or iridium, thread it into a glass capillary about a millimeter in diameter, and then put it on a flame and turn it over until the glass softens, and the glass tube is elongated and very thin like an optical fiber, and the result is a hard electrode with glass insulation, the tip of which may be only a few tens of microns in size.

These neurologists implant these hard electrodes through the skull to the surface of the brain — above or below the dura mater, or on the surface and inside of the cerebral cortex.

Scientists are now developing a patch-clamp technique in which the tip of the electrode is removed and a small glass straw is left into which a portion of the neuronal cell membrane is sucked into the glass tube for more accurate measurements.

Another extreme is that the electrodes pierce the cell membrane and completely enter the neuron, this method is called sharp electrode recording, which has a very high probability of destroying the neuronal cells, and the pierced neurons cannot survive for a long time.

These electrodes that invade the brain are very easy to damage the brain, and they will be repelled by the brain, which is very dangerous, so it has also received a lot of opposition abroad.

This method is very crude and barbaric, as primitive as the removal of diseased brains forty or fifty years ago in order to treat brain diseases.

Due to the understanding of the brain and the existing electrode hardware are very primitive, foreign related research mainly focuses on making some simple interfaces, such as the motor cortex and visual cortex, and no small breakthroughs have been made in this regard, such as the emergence of cochlear implants and retinal prostheses.

To date, more than 100,000 deaf people have used cochlear implants, more than half of whom are children.

Cochlear implants are now used worldwide as a routine treatment for severe to total deafness, and cochlear implants are currently the most successful biomedical engineering device.

Although retinal implants can now repair vision in a similar way to cochlear implants repairing hearing, transmitting information to nerves in the form of electronic impulses, it is a more complex brain-computer interface than cochlear implants.

Although retinal prostheses are not yet on the market, some foreign companies have made great breakthroughs in research and development.

In my memory, the first FDA-approved retinal prosthesis was launched in 2011, this retinal prosthesis with 60 sensors, compared to a real retina with about 1 million neurons, although this prosthesis may seem rough, but at least it can allow the blind to see the blurred edges, shapes, and light and dark changes of objects, which is better than not seeing nothing.

And retinal prostheses with 600 to 1,000 electrodes are enough to provide vision for reading and face discrimination.

At present, the brain-computer interface R&D department has gathered almost several scientists from the fields of brain science, electrochemistry, biology, microelectronic engineering, etc., and cooperates with several medical universities in China to develop brain-computer interface technology for the motor cortex and visual cortex.

In particular, in the research and development of motor neurons, several universities in China have developed this area, and the brain-computer interface department has made some progress in the bioelectric signals of motor neurons, and has now analyzed the bioelectric signals of some neurons to control the muscles of the limbs to make movements.

The sensors used by the second-generation exoskeleton robot now rely on the movement control of the exoskeleton robot by feeling the changes in the user's muscle pressure, and can still be used if the user's muscles are not atrophied, but there is some powerlessness in the face of amyotrophic spinal cord sclerosis and high amputation.

The R&D team also proposed to Yang Jie to develop a new type of sensor technology, so that the sensor can directly detect the electrical signals sent by the motor neuron of the user's spinal cord, which can be more accurate and more complex than relying solely on muscle twitching to control, which can make the exoskeleton robot closer to intuitive control in use, and the usefulness of the disabled will be greater, and the practicability will be greatly improved.

Yang Jie also approved the R&D project, the headquarters allocated 15 million US dollars for research and development, the project research and development time is three years, requiring the team to come up with the third generation of exoskeleton robots within three years, and the headquarters also allocated 8 million US dollars for the research and development center of the company in Huaxia to develop a new type of microelectromechanical sensor.

The Robotics Research Laboratory also has a project on a bionic robotic arm and a research and development project for retinal prostheses.

Yang Jie also hopes that the R&D team can make the bionic robotic arm complete actions such as moving the prosthetic elbow joint, swinging the prosthetic wrist, opening and closing the palm in more than ten years, so as to realize most of the basic functions of the real arm and wrist.

To be able to do this is not only to help people with disabilities, but to enable robots to start entering more industries.

Today's robots can only be limited to some specific industries, such as automobile production lines, and manipulators still can't be as flexible as humans.

The universal bionic manipulator is what Yang Jie wants to do.