Studies show promising alternatives to personalized deep brain stimulation

Millions of patients suffering from neurological and psychiatric disorders such as depression, addiction, and chronic pain are refractory to treatment. In fact, about 30% of all patients with major depression do not respond to drug or psychotherapy at all.

Simply put, many traditional treatments for these disorders may have reached their limits. Where are you going from here?

Studies published in Nature Biomedical Engineering It paves the way for personalized deep brain stimulation, a promising alternative, led by Maryam Shanechi, Andrew and Erna Viterbi Early Career Chair of the USC Viterbi School of Engineering.

This study represents a major step forward in achieving new treatments for many neurological and psychiatric disorders.

Until now, the challenge of personalized deep brain stimulation has been the human brain itself. Psychiatric disorders can appear differently in each patient’s brain. Similarly, each patient’s brain activity and its symptoms differ greatly in whether and how they respond to the stimulus.

This makes it difficult to know the effect of the stimulus on a particular patient and how to change the stimulus dose (that is, its amplitude or frequency) over time to suit the patient’s needs.

Shanechi and her team find a way to predict how electrical stimulation affects an individual’s brain activity across multiple brain regions by developing new stimulus waveforms and creating new machine learning models. I did. They collaborated with Bijan Pesaran, a professor of neuroscience at New York University, to demonstrate the success of the model in real-world brain stimulation experiments.

To achieve this, they designed two tools. A new electrical stimulation wave for mapping brain activity. A new machine learning technology that learns maps from brain data collected during stimulation.

Our waves, which randomly change amplitude and frequency over time, have made it possible to see and predict how the brain responds to a wide range of stimulus doses. “

Maryam Shanechi, Andrew and Erna Viterbi Early Career Chair, Electrical and Computer Engineering, University of Southern California Viterbi Engineering

Just as a skeleton key can unlock a door, waves can be applied to an individual’s brain and provide a personalized map of how they respond to stimuli. To test their hypothesis, researchers applied their waves to four different areas of the brain. In each case, for the first time, they were able to predict the outcome of brain activity across multiple regions.

What this means is that physicians may be able to quickly personalize the “dose” of deep brain stimulation in real time on a case-by-case basis by varying the amplitude and frequency of stimulation.

Think of it as a brain-stimulating version that increases or decreases the number of milligrams of the pill. The consequences are enormous for people suffering from mental disorders such as treatment-resistant depression and anxiety.

Shanechi and her team previously developed machine learning techniques to decipher symptoms of mental disorders such as mood from brain activity. Now, with a new ability to more accurately predict on an individual basis how stimuli affect brain activity, they are trying to combine their findings towards personalized treatment of mental illness.

“The combination of these two boxes is a closed-loop brain machine that adjusts the dose of electrical stimulation therapy by tracking symptoms in real time based on brain activity and predicting how changes in stimulation will change. I want to build an interface. Activities, and therefore these symptoms. “