Maryam Shanechi, Sawchuk Chair of Electrical and Computer Engineering and founding director of the USC Neurotechnology Center, and her team have developed a new AI algorithm that can tease apart brain patterns related to a particular behavior. This work, which may improve brain-computer interfaces and uncover new brain patterns, has been published in the journal Neuroscience of nature.
As you read this story, your brain is engaged in multiple behaviors.
Maybe you are moving your arm to grab a cup of coffee, while reading the article out loud to your colleague and feeling a little hungry. All these different behaviors, such as arm movements, speech, and different internal states such as hunger, are simultaneously encoded in your brain. This simultaneous encoding gives rise to very complex and confusing patterns in the brain’s electrical activity. Therefore, a big challenge is to dissociate those brain patterns that encode a particular behavior, such as arm movement, from all the other brain patterns.
For example, this dissociation is key to developing brain-computer interfaces that aim to restore movement in paralyzed patients. When they think about making a movement, these patients cannot communicate their thoughts to their muscles. To restore function in these patients, brain-computer interfaces decode the planned movement directly from their brain activity and translate it into the movement of an external device, such as a robotic arm or a computer cursor.
Shanechi and his former PhD student, Omid Sani, who is now a research associate in his lab, developed a new AI algorithm that addresses this challenge. The algorithm is called DPAD, for “Dissociative Prioritized Analysis of Dynamics.”
“Our AI algorithm, called DPAD, dissociates brain patterns that encode a particular behavior of interest, such as arm movement, from all other brain patterns occurring at the same time,” Shanechi said. “This allows us to decode brain activity movements more accurately than previous methods, which can improve brain-computer interfaces. In addition, our method can also uncover new patterns in the brain that might otherwise go unnoticed.”
“A key element of the AI algorithm is to first look for brain patterns related to the behavior of interest and learn these patterns in priority during the training of a deep neural network,” Sani added. “After doing this, the algorithm can later learn all the remaining patterns so that they do not mask or confuse the behavior-related patterns. Moreover, the use of neural networks provides a wide flexibility in terms of the types of brain patterns the algorithm can describe.”
In addition to movement, this algorithm has the flexibility to be used in the future to decode mental states such as pain or depressed mood. This can help to better treat mental health conditions by tracking a patient’s symptomatic states as feedback to precisely tailor their therapies to their needs.
“We are very excited to develop and demonstrate extensions of our method that can track symptomatic states in mental illness,” Shanechi said. “Doing so could lead to brain-computer interfaces not only for movement disorders and paralysis, but also for mental illness.”