Artificial intelligence may one day treat neurological conditions like epilepsy and Parkinson's disease.
According to an article in Interesting Engineering, a team of University of Toronto researchers integrated microelectronics and artificial intelligence to create unique, safe technology.
Researchers hope to build on work on deep learning for neural implants used to detect seizures, too. Beyond epileptic episodes, researchers said that the team's work might be applied in a wide range of therapeutic applications and medical procedures.
They hope to apply the technique to treat a range of brain conditions that afflict up to one billion people worldwide. They expect those novel therapeutic treatments may be developed for individuals with dementia, chronic pain, Alzheimer's disease, and depression in addition to researching the impact on epilepsy and Parkinson's disease.
Artificial Intelligence Might Be Able To Treat Parkinson's Disease and Epilepsy
According to a study titled "Edge deep learning for neural implants: a case study of seizure detection and prediction," the team aims to produce tiny silicone chips with brain implants similarly embedded in them to how current computer chips are made.
The neural implant can aid in this by delivering electrical stimulation since neurons interact through electrical impulses.
The stimulation will restore neurons to normal when a patient suffers seizures or tremors.
It won't be as simple even though it is just an on/off switch. In actuality, experts are still baffled by the project's intricacy.
Because of this, they think artificial intelligence has the potential to one day be a practical therapeutic choice.
Additionally, they aim to mitigate the harmful consequences of excessive brain stimulation.
The Engineered Arts Ameca humanoid robot with artificial intelligence is demonstrated during the Consumer Electronics Show (CES) on January 5, 2022, in Las Vegas, Nevada.
ALSO READ: Developing AI Technology Enabling Robots to Communicate With Some Animal Species
Neural Implants
The CMOS technology used by the researchers allows them to reduce the size of the gadget and improve power use.
Therefore, it helps to reduce risks associated with long-term usage and surgical implantation of the neural implant.
To create the best prototype, the team has experimented with several strategies and techniques, like high-precision electrical stimulation with charge balancing.
Researchers used deep learning, which uses artificial neural networks (per Mathworks). It also aids in the discovery of hidden biomarkers utilizing a combination of algorithms that learn and retrieve deep-level information.
This allows scientists to trigger brain implants using specific biomarkers. They can continue to utilize the stimulation since much of the guessing is removed.
Artificial intelligence can only turn on brain implants when absolutely necessary.
One issue with this is the cost of computation. It would be difficult to implement this technology using deep learning models.
In order to reduce processing costs, the researchers created methods for training the models based on each patient's health.
RELATED ARTICLE: New AI Technology Detects Invisible Patterns of MRI to Accurately Predict If and When a Person Dies From Cardiac Arrest, Study Claims
Check out more news and information on Artificial Intelligence in Science Times.
