Brain disorders like Parkinson's disease and epilepsy may soon be treated by artificial intelligence, as reported by Interesting Engineering. A team of researchers at the University of Toronto combined artificial intelligence and microelectronics to create innovative technology that is safe and effective.
Neural Implants
The team's aim is to incorporate neural implants into miniature silicone chips the same way it does to manufacture chips used in our computers today. Since neurons communicate with each other through electrical signals, the neural implant can help with this by producing electric stimulation.
When a patient has seizures or tremors, the stimulation will try to restore neurons to normal. It's basically an on and off switch; but it won't be that simple. In fact, the researchers are still trying to understand how complex the project is.
That's why they see the potential of artificial intelligence as an effective treatment option in the future. They also aim to help minimize adverse reactions to the excessive brain stimulation.
Also Read: Research Suggests Human Brain Is 30 Times As Powerful As The Best Supercomputers
Complementary Metal-Oxide Semiconductor (CMOS)
The technology is called CMOS that enables the researchers to reduce the size of the device and for power utilization. As a result, it helps reduce risks connected with the neural implant's surgical procedure and long-standing usage.
The team has tried various strategies and techniques to create the best prototype, such as high-precision electrical stimulation with charge balancing.
As for the AI used, they used deep learning that uses artificial neural networks. Through a compilation of algorithms that learn and extract deep-level information, it also helps identify hidden biomarkers.
With this, researchers are able to activate the neural implants based on biomarkers. This takes out plenty of the guesswork and enables them to use the stimulation continuously.
Artificial intelligence can activate neural implants at an optimal time and only when needed.
The Challenge
One challenge to this is the computational cost. Deep-learning models would make it difficult to incorporate this type of technology; and a way to reduce the computational costs, the researchers created methods for training the models based on each patient's condition.
The Future of Neural Implants
The researchers believe that their miniaturized neural implants will make it easier to help patients, especially those who don't respond well to medication. The neural implants will keep the patients' neurological processes functioning optimally.
While it may not be a "cure" to brain disorders yet, it will help improve the quality of life for patients. It will also help minimize the risk of side-effects that come with medication. Neural implants can also be used to improve human condition.
This is only the beginning. The project is still in the works, so there's plenty of time to incorporate AI. But since brain disorders are complex in nature and require continuous treatment, this could be an effective way to treat it.
Related Article: Harvard To Develop AI That Works As Fast As Human Brain
This article is owned by Tech Times
Written by April Fowell
