-
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
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Written byWang CongEditorWang Duoyu TypesettingShui Chengwen
Neural Implants can help treat brain diseases such as Parkinson's disease and epilepsy by
directly regulating abnormal neuronal activity.
Liu Xilin of the University of Toronto is combining microelectronics and artificial intelligence (AI) to make this emerging technology both safe and smart
.
Liu Xilin, an assistant professor in the School of Applied Science and Engineering at the University of Toronto, said that communication between neurons is partly carried out through electrical signals, and therapeutic neural implants are capable of generating electrical stimulation, like the brain's pacemaker
.
In case of tremors or seizures, neurons are restored to their normal state
by electrical stimulation.
Turning neural networks off and on through electrical stimulation, like restarting a computer, is of course not that simple, and scientists don't yet fully understand how it works
.
Liu Xilin's team integrated the neural implants into micro-silicon chips, the same
process used in today's computers and smartphones.
This technique, known as CMOS (complementary metal-oxide semiconductor), is able to reduce the physical size and power consumption of the device, thereby minimizing the risks associated with the initial surgical procedure and long-term use of the
implant.
Liu Xilin said that the team has developed many new microelectronic design techniques, such as high-precision electrical stimulation with charge balance, to try to solve this problem
from many different angles.
Liu Xilin is part of the Centre for Neurotechnology at CRUIA, a collaboration between the University of Toronto and Toronto General Hospital that brings together electrical and computer engineers, neuroscientists, data and materials scientists, and clinicians
.
Together, they research ways to improve brain health and develop alternative treatments, especially for those who
don't respond well to current medications.
In a recent project, Liu Xilin's team tried to harness the power of artificial intelligence (AI) to maximize the clinical efficacy of implants and minimize
the adverse effects of overstimulation.
On Liu Xilin's fingertips is a 3×3mm neural implant chip prototype
team using deep learning AI algorithms, which are trained to extract deep information
when faced with new data.
Deep learning algorithms have proven to be particularly powerful at identifying hidden biomarkers that are often overlooked by traditional methods, and outperform traditional algorithms
.
Liu Xilin says most existing neural implants generate electrical stimulation at a constant rate, regardless of the
patient's actual condition.
With deep learning algorithms, neural implants can be activated at the optimal time point and only when necessary
.
However, the high computational cost of deep learning models makes their integration a challenge, especially considering that all processing must run
locally in the implant.
Cloud technology can provide more processing power, but it also has an obvious drawback, when a patient enters an elevator or plane, the implant can fail due to the loss of communication service
.
To reduce computational costs, Liu Xilin's team developed techniques
to train and optimize models for each patient's situation.
A recent study showed that detecting seizures through deep learning in low-power neural implants is comparable to state-of-the-art algorithms running in high-performance computers.
The work was published in the Journal of Neural Engineering in 2021
.
As many as 1 billion people worldwide suffer from various brain diseases, and this technology has a wide range
of clinical applications in addition to helping with the detection and treatment of epilepsy.
There is so much going on in the brain that it takes a range of experts to understand and provide solutions to these brain diseases, which are only going to become more common
as human life expectancy increases.
Liu Xilin said that in addition to epilepsy, future goals include chronic pain, depression and dementia
.
Liu Xilin said that impaired sleep is linked to Alzheimer's disease, and that many people have varying degrees of sleep disorders, and the team is working on closed-loop neuromodulation techniques to help Alzheimer's patients
by strengthening or inhibiting certain brain rhythms to improve sleep quality.
Open reprint welcome to forward to Moments and WeChat groups
Neural Implants can help treat brain diseases such as Parkinson's disease and epilepsy by
directly regulating abnormal neuronal activity.
Liu Xilin of the University of Toronto is combining microelectronics and artificial intelligence (AI) to make this emerging technology both safe and smart
.
Liu Xilin, an assistant professor in the School of Applied Science and Engineering at the University of Toronto, said that communication between neurons is partly carried out through electrical signals, and therapeutic neural implants are capable of generating electrical stimulation, like the brain's pacemaker
.
In case of tremors or seizures, neurons are restored to their normal state
by electrical stimulation.
Turning neural networks off and on through electrical stimulation, like restarting a computer, is of course not that simple, and scientists don't yet fully understand how it works
.
Liu Xilin's team integrated the neural implants into micro-silicon chips, the same
process used in today's computers and smartphones.
This technique, known as CMOS (complementary metal-oxide semiconductor), is able to reduce the physical size and power consumption of the device, thereby minimizing the risks associated with the initial surgical procedure and long-term use of the
implant.
Liu Xilin said that the team has developed many new microelectronic design techniques, such as high-precision electrical stimulation with charge balance, to try to solve this problem
from many different angles.
Liu Xilin is part of the Centre for Neurotechnology at CRUIA, a collaboration between the University of Toronto and Toronto General Hospital that brings together electrical and computer engineers, neuroscientists, data and materials scientists, and clinicians
.
Together, they research ways to improve brain health and develop alternative treatments, especially for those who
don't respond well to current medications.
In a recent project, Liu Xilin's team tried to harness the power of artificial intelligence (AI) to maximize the clinical efficacy of implants and minimize
the adverse effects of overstimulation.
On Liu Xilin's fingertips is a 3×3mm neural implant chip prototype
team using deep learning AI algorithms, which are trained to extract deep information
when faced with new data.
Deep learning algorithms have proven to be particularly powerful at identifying hidden biomarkers that are often overlooked by traditional methods, and outperform traditional algorithms
.
Liu Xilin says most existing neural implants generate electrical stimulation at a constant rate, regardless of the
patient's actual condition.
With deep learning algorithms, neural implants can be activated at the optimal time point and only when necessary
.
However, the high computational cost of deep learning models makes their integration a challenge, especially considering that all processing must run
locally in the implant.
Cloud technology can provide more processing power, but it also has an obvious drawback, when a patient enters an elevator or plane, the implant can fail due to the loss of communication service
.
To reduce computational costs, Liu Xilin's team developed techniques
to train and optimize models for each patient's situation.
A recent study showed that detecting seizures through deep learning in low-power neural implants is comparable to state-of-the-art algorithms running in high-performance computers.
The work was published in the Journal of Neural Engineering in 2021
.
As many as 1 billion people worldwide suffer from various brain diseases, and this technology has a wide range
of clinical applications in addition to helping with the detection and treatment of epilepsy.
There is so much going on in the brain that it takes a range of experts to understand and provide solutions to these brain diseases, which are only going to become more common
as human life expectancy increases.
Liu Xilin said that in addition to epilepsy, future goals include chronic pain, depression and dementia
.
Liu Xilin said that impaired sleep is linked to Alzheimer's disease, and that many people have varying degrees of sleep disorders, and the team is working on closed-loop neuromodulation techniques to help Alzheimer's patients
by strengthening or inhibiting certain brain rhythms to improve sleep quality.
Open reprint welcome to forward to Moments and WeChat groups
