-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
- Cosmetic Ingredient
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Author | nagashi comment | Edited by Dr.
Shuqiang Lu, founder of Brainland Technology | Wang Cong Once upon a time, the connection between human brain and machine was just a concept that appeared in science fiction and film and television, but now, with the development of brain-computer interface technology (BMI) , This sci-fi concept is being written into reality step by step.
In the foreseeable future, brain-computer interface technology will have abundant application space in the fields of medical care, education, entertainment, consumption, and smart cars.
In February 2020, Neuralink, a brain-computer interface startup under Tesla CEO Elon Musk, connected a computer chip to the monkey's brain, allowing the monkey to control the table tennis game through its own ideas.
In August of the same year, Neuralink once again launched a new device-Link V0.
9, and successfully implanted this 23mm diameter chip into the pig's brain, realizing real-time neural signal reading and writing.
But on a practical level, as a transformative technology, the development of brain-computer interfaces still faces many challenges from technology, policy, and ethics.
Among them, the most prominent point is whether the implanted chip will cause harm to the human body, the possible rejection, infection reaction, and the heat dissipation and other instability factors caused by the operation of the chip.
From this point of view, there are still many problems that need to be solved urgently in the current chip-implanted brain-computer interface technology.
Therefore, it is particularly important to develop a minimally invasive, safer and more accurate brain-computer interface technology.
On March 22, 2021, research institutions such as Chen Tianqiaoluoqianqian Brain-Computer Interface Center, California Institute of Technology, French National Institute of Health and Medicine (INSERM) and other research institutes published the title: Single-trial decoding of movement in the Cell sub-Journal Neuron.
A research paper on intentions using functional ultrasound neuroimaging.
This latest research uses minimally invasive functional ultrasound (fUS) neuroimaging technology to read the monkey’s brain activity and predict its next eye or hand movement.
This information can generate commands for a robotic arm or a computer cursor.
It will provide paralyzed people with a new method of controlling prosthetics without the need for devices that invade the brain.
Functional ultrasound predicts the movement of rhesus monkeys For a long time, ultrasound has been widely used in human medicine, detection and industry, such as B-ultrasound during pregnancy and sonar detection.
About ten years ago, scientists invented a method for brain imaging using ultrasound.
This method is called functional ultrasound imaging.
This technology uses a wide, flat sound rather than a narrow sound beam, and can capture a large area of sound faster than traditional ultrasound.
Similar to functional magnetic resonance imaging (fMRI), functional ultrasound can measure changes in blood flow to show when neurons are active and consume energy.
More importantly, the resolution of images generated by functional ultrasound imaging technology is much higher than that of fMRI and does not require large-scale scanning equipment.
In this study, the researchers inserted some small ultrasonic transducers into the skulls of two rhesus monkeys, the size and shape of which are similar to dominoes.
The device is connected to a computer via a data cable, and then enters the posterior parietal cortex area of the brain through functional ultrasound, which is a brain area that is important for spatial perception, multi-sensory integration, and motor planning.
Functional ultrasound scanning area.
Then, the researchers trained these monkeys to focus their eyes on a small spot in the center of the screen, and the second spot would flash across the screen briefly, then the center spot disappeared, and the monkeys would look away Go to the place where the second point has recently flashed.
In another set of experiments, these monkeys were trained to reach out and move the joystick to point to that point.
At the same time, a computer algorithm converts the ultrasound data into a prediction of the monkey's intent.
The algorithm can determine when the monkey is ready to move, and whether they are planning eye movements or arm extensions.
The researchers said that through this algorithm, they can predict whether the monkey's movement is to the left or right.
The accuracy of eye movements is about 78%, and the accuracy of stretching is 89%.
A safe and easy-to-accept minimally invasive technology, but still needs improvement.
More importantly, this technology is minimally invasive.
Although a small piece of skull still needs to be removed, it is not the same as an implant that directly reads the electrical activity of neurons Unlike the electrode, it does not need to open the protective membrane of the brain.
And, functional ultrasound can read information from deep in the brain without penetrating tissue.
Single decoding of expected direction of movement Of course, measuring neural activity from a distance means sacrificing some speed and accuracy.
Professor Mikhail Shapiro from California Institute of Technology, one of the corresponding authors of this study, said: Functional ultrasound provides “a not-so-direct signal”.
Therefore, it is very important to accurately interpret the information in the ultrasound image.
At this point, it is necessary to perform an average prediction on a signal with a long period of time or multiple motions.
Not only that, computer algorithms must be as convenient for users as possible, which requires a lot of attempts to study how the brain coordinates movement.
For example, a user of a robotic arm only needs to consider the movement they want once to move the arm, and does not need to do more actions in order for the algorithm to interpret their intentions.
The influence of spatial resolution, time window and average power Doppler intensity is worth noting that compared with chip implanted brain-computer interface technology, functional ultrasound still has many things to be improved in decoding speed and motion interpretation ability.
local. Nowadays, chip-implanted brain-computer interface technology can decode multiple directions of the arm, not just the left and right directions.
In response, Emilie Macé, a neuroscientist at the Max Planck Institute for Neurobiology, added: “Because blood flow signals are slower than electrical signals, speed is an inherent limitation of functional ultrasound.
Researchers need about 2 seconds.
To decode the monkey’s exercise plan.
"Of course, there is still room for improvement in this technology, such as collecting more information by imaging 3D blocks of brain tissue instead of planar imaging.
This technology has not yet reached its full potential.
.
Conclusion Dr.
Shuqiang Lu, founder of Brain-Computer Interaction, a startup company in the field of brain-computer interaction, told Biological World: "This technology requires a craniotomy, but compared to traditional invasive solutions, it has less brain damage.
The
collected signals are mainly used for disease diagnosis.
, Interactive control may be more difficult, because the output is ultrasound image signals, not brain electrical signals.
"
Dr.
Lu Shuqiang also said: "If you can further use artificial intelligence algorithms, you can achieve higher-order tasks.
If you can achieve three-dimensional brain ultrasound, it is much more data-rich than other current tools, and the application of imagination is also greater.
"
All in all, the research jointly led by Professor Richard Andersen and Professor Mikhail Shapiro used functional ultrasound neuroimaging to record changes in the cerebral blood volume of monkeys with a resolution of 100 μm, and to predict their behavior and movement, including the accuracy of eye movements.
About 78%, and the accuracy of stretching exercises was 89%.
This research result has taken a critical step towards the development of neural recording and brain-computer interface tools with small trauma, high resolution, and scalability.
The innovation of science and technology brings endless possibilities to mankind.
Perhaps in the foreseeable future, the real-time interaction between human brain and machine is no longer a science fiction concept, it will profoundly change our lives! Link to the paper: Open for reprint
Shuqiang Lu, founder of Brainland Technology | Wang Cong Once upon a time, the connection between human brain and machine was just a concept that appeared in science fiction and film and television, but now, with the development of brain-computer interface technology (BMI) , This sci-fi concept is being written into reality step by step.
In the foreseeable future, brain-computer interface technology will have abundant application space in the fields of medical care, education, entertainment, consumption, and smart cars.
In February 2020, Neuralink, a brain-computer interface startup under Tesla CEO Elon Musk, connected a computer chip to the monkey's brain, allowing the monkey to control the table tennis game through its own ideas.
In August of the same year, Neuralink once again launched a new device-Link V0.
9, and successfully implanted this 23mm diameter chip into the pig's brain, realizing real-time neural signal reading and writing.
But on a practical level, as a transformative technology, the development of brain-computer interfaces still faces many challenges from technology, policy, and ethics.
Among them, the most prominent point is whether the implanted chip will cause harm to the human body, the possible rejection, infection reaction, and the heat dissipation and other instability factors caused by the operation of the chip.
From this point of view, there are still many problems that need to be solved urgently in the current chip-implanted brain-computer interface technology.
Therefore, it is particularly important to develop a minimally invasive, safer and more accurate brain-computer interface technology.
On March 22, 2021, research institutions such as Chen Tianqiaoluoqianqian Brain-Computer Interface Center, California Institute of Technology, French National Institute of Health and Medicine (INSERM) and other research institutes published the title: Single-trial decoding of movement in the Cell sub-Journal Neuron.
A research paper on intentions using functional ultrasound neuroimaging.
This latest research uses minimally invasive functional ultrasound (fUS) neuroimaging technology to read the monkey’s brain activity and predict its next eye or hand movement.
This information can generate commands for a robotic arm or a computer cursor.
It will provide paralyzed people with a new method of controlling prosthetics without the need for devices that invade the brain.
Functional ultrasound predicts the movement of rhesus monkeys For a long time, ultrasound has been widely used in human medicine, detection and industry, such as B-ultrasound during pregnancy and sonar detection.
About ten years ago, scientists invented a method for brain imaging using ultrasound.
This method is called functional ultrasound imaging.
This technology uses a wide, flat sound rather than a narrow sound beam, and can capture a large area of sound faster than traditional ultrasound.
Similar to functional magnetic resonance imaging (fMRI), functional ultrasound can measure changes in blood flow to show when neurons are active and consume energy.
More importantly, the resolution of images generated by functional ultrasound imaging technology is much higher than that of fMRI and does not require large-scale scanning equipment.
In this study, the researchers inserted some small ultrasonic transducers into the skulls of two rhesus monkeys, the size and shape of which are similar to dominoes.
The device is connected to a computer via a data cable, and then enters the posterior parietal cortex area of the brain through functional ultrasound, which is a brain area that is important for spatial perception, multi-sensory integration, and motor planning.
Functional ultrasound scanning area.
Then, the researchers trained these monkeys to focus their eyes on a small spot in the center of the screen, and the second spot would flash across the screen briefly, then the center spot disappeared, and the monkeys would look away Go to the place where the second point has recently flashed.
In another set of experiments, these monkeys were trained to reach out and move the joystick to point to that point.
At the same time, a computer algorithm converts the ultrasound data into a prediction of the monkey's intent.
The algorithm can determine when the monkey is ready to move, and whether they are planning eye movements or arm extensions.
The researchers said that through this algorithm, they can predict whether the monkey's movement is to the left or right.
The accuracy of eye movements is about 78%, and the accuracy of stretching is 89%.
A safe and easy-to-accept minimally invasive technology, but still needs improvement.
More importantly, this technology is minimally invasive.
Although a small piece of skull still needs to be removed, it is not the same as an implant that directly reads the electrical activity of neurons Unlike the electrode, it does not need to open the protective membrane of the brain.
And, functional ultrasound can read information from deep in the brain without penetrating tissue.
Single decoding of expected direction of movement Of course, measuring neural activity from a distance means sacrificing some speed and accuracy.
Professor Mikhail Shapiro from California Institute of Technology, one of the corresponding authors of this study, said: Functional ultrasound provides “a not-so-direct signal”.
Therefore, it is very important to accurately interpret the information in the ultrasound image.
At this point, it is necessary to perform an average prediction on a signal with a long period of time or multiple motions.
Not only that, computer algorithms must be as convenient for users as possible, which requires a lot of attempts to study how the brain coordinates movement.
For example, a user of a robotic arm only needs to consider the movement they want once to move the arm, and does not need to do more actions in order for the algorithm to interpret their intentions.
The influence of spatial resolution, time window and average power Doppler intensity is worth noting that compared with chip implanted brain-computer interface technology, functional ultrasound still has many things to be improved in decoding speed and motion interpretation ability.
local. Nowadays, chip-implanted brain-computer interface technology can decode multiple directions of the arm, not just the left and right directions.
In response, Emilie Macé, a neuroscientist at the Max Planck Institute for Neurobiology, added: “Because blood flow signals are slower than electrical signals, speed is an inherent limitation of functional ultrasound.
Researchers need about 2 seconds.
To decode the monkey’s exercise plan.
"Of course, there is still room for improvement in this technology, such as collecting more information by imaging 3D blocks of brain tissue instead of planar imaging.
This technology has not yet reached its full potential.
.
Conclusion Dr.
Shuqiang Lu, founder of Brain-Computer Interaction, a startup company in the field of brain-computer interaction, told Biological World: "This technology requires a craniotomy, but compared to traditional invasive solutions, it has less brain damage.
The
collected signals are mainly used for disease diagnosis.
, Interactive control may be more difficult, because the output is ultrasound image signals, not brain electrical signals.
"
Dr.
Lu Shuqiang also said: "If you can further use artificial intelligence algorithms, you can achieve higher-order tasks.
If you can achieve three-dimensional brain ultrasound, it is much more data-rich than other current tools, and the application of imagination is also greater.
"
All in all, the research jointly led by Professor Richard Andersen and Professor Mikhail Shapiro used functional ultrasound neuroimaging to record changes in the cerebral blood volume of monkeys with a resolution of 100 μm, and to predict their behavior and movement, including the accuracy of eye movements.
About 78%, and the accuracy of stretching exercises was 89%.
This research result has taken a critical step towards the development of neural recording and brain-computer interface tools with small trauma, high resolution, and scalability.
The innovation of science and technology brings endless possibilities to mankind.
Perhaps in the foreseeable future, the real-time interaction between human brain and machine is no longer a science fiction concept, it will profoundly change our lives! Link to the paper: Open for reprint
