Artificial Intelligence continues to break new ground within the fascinating realm of neuroscience. Recently, researchers have utilized AI to distinguish specific patterns of brain activity linked to certain behaviors. This development in brain-computer interfaces (BCIs) technology and discoveries of unique brain patterns could revolutionize how we understand and interact with the brain.
The research, conducted by Maryam Shanechi, the Sawchuk Chair in Electrical and Computer Engineering, founding director of the USC Center for Neurotechnology, along her team, yielded an innovative AI algorithm that can separate distinct brain patterns related to a specific behavior. The findings have been published in the widely renowned journal Nature Neuroscience.
The algorithm manages to discern and prioritize specific brain activities that encode a variety of behaviors. For instance, while you read this article, your brain may indulge in multiple behaviors concurrently, like moving your arm, experiencing hunger, or reading aloud. The brain codes all these behaviors at once, producing intricate and intertwined patterns in the brain's electrical activity. The challenge lies in decoupling a given behavior, such as arm movement, from the overarching brain patterns.
This decoupling process is crucial in the development of brain-computer interfaces aimed at restoring function for patients impaired by paralysis. The patients, unable to communicate their intent to move to their muscles, can now, with the help of BCIs, decode the planned movement right from their brain activity. It translates this decoded information into action by moving an external device, like a robotic arm or computer cursor.
Shanechi, in collaboration with her former Ph.D. student, Omid Sani, who is currently a research associate in her lab, developed an AI algorithm called DPAD (Dissociative Prioritized Analysis of Dynamics) that can tackle this complexity. It dissects the brain activities related to a particular behavior from all other co-existing patterns, allowing for a more accurate decoding of movements from brain activity than previous methods.
Sani further explains, "a critical component of our AI algorithm is to first identify brain patterns related to the behavior of interest and learn these patterns with priority during the training of a deep neural network. Subsequently, the algorithm learns all remaining patterns to ensure they do not cloud or convolute the behavior-related patterns."
Flexibility is another remarkable feature of this algorithm. It has the potential to decode mental states like pain, or depressed mood in the future, thereby aiding in the treatment of various mental health conditions by enabling real-time tracking of a patient's symptoms. "We are excited at the prospect of extending our method that can track symptom states in mental health conditions leading to BCIs adapted not only for movement disorders and paralysis but also for mental health conditions," Shanechi declares.
Disclaimer: The above article was written with the assistance of AI. The original sources can be found on ScienceDaily.
