Science and artificial intelligence were combined at the University of Medicine at South Carolina in a study that could lead to individual repeat cranial magnetic stimulation (RTM) for smokers who want to quit.
We hope to improve the effectiveness and specificity of RTMS and reduce side effects. ”
Xingbao Li, MD, Research Leader
He is an associate professor in the Department of Psychiatry and Behavioral Sciences who conducted extensive research on TMS.
His team presented the results in Journal Brain Connectivity.
TMS uses electromagnetic pulses to influence brain activity and may be best known for its role in the treatment of depression and obsessive-compulsive disorder. Side effects include discomfort and headaches at the stimulating site.
TMS is also approved by the Food and Drug Administration for smoking. In South Carolina it was the first place to provide TMS to smokers. Studies have shown that multiple sessions of RTMS, particularly on the left dorsolateral prefrontal cortex of the brain, can reduce craving and tobacco consumption.
The new MUSC research is even more targeted than that. Using a form of AI called MachineRearning, we analyze images from the brain's neural network to see if we can predict which smokers could potentially benefit from multiple sessions of RTMS (also known as iterative TMS).
To that end, researchers used functional magnetic resonance imaging, or fMRI, to detect changes in blood flow to measure brain activity. They viewed neural networks when participants were in rest, closed their eyes and relaxed, and exposed to smoking photos.
That analysis found that one neural network stands out: the Salience network. Filter the information to determine what is prominent or important to focus on. In this study, connectivity of the Salience network was the best predictor of RTMS effectiveness.
“This study provides a roadmap for extending personalized RTMs and building fMRI and multimodal biomarker pipelines. This method can also be used for other substance use disorders,” Li said.
“Historical research focuses on the reward network of cigarette smokers,” he continues, referring to the parts of the brain that are involved in motivation and joy.
“We were surprised that the prominent networks play a very important role in smoking behavior. This makes the salience network a mechanical bridge between RTMS neuroregulation and the success of smoking bans.”
They found the bridge with the help of machine learning. In machine learning, computers analyze and learn from data without being programmed. Use algorithms that can find and adapt statistical patterns. This allows researchers to automate that part of their work and improve accuracy.
In this case, machine learning analyzed data collected during previous MUSC studies on smokers' TMS.
Here's how previous research was set up: The researchers recruited 42 people who wanted to stop smoking. They were divided into two groups. One group got the actual TMS. The other one got a fake TMS that felt like the real thing. They all spent a minute and a half before each TMS session interacting with cigarettes, ashtrays and more. And then I watched a video of smoking, between TMS, real or fake. There were 10 sessions per person over two weeks.
Finally, the researchers found participants who obtained “significantly fewer tobacco cigarettes per day during two weeks of treatment,” and were more likely to quit by the target date and had a low craving for cigarettes.
Li said that the new research could be built on the findings, thanks to fMRI scans, which was also part of the work. “Using machine learning to identify individual dysfunctional brain networks and applying RTMS to dysfunctional networks allows you to choose who or who prefers to use RTMS to help you stop smoking.”
This study was supported by grants from the National Institutes of Health. The authors did not report any conflict of interest. The research team includes Kevin Caulfield, Ph.D. Included. Dr. Andrew Chen; Dr. Christopher McMahan; Karen Hartwell, MD; Kathleen Brady, MD; Dr. Mark George, Maryland
Li said their relatively small studies will serve as the basis for large-scale studies to further explore targeted TMSs for smokers. “This shows that MUSC researchers can use novel and impactful techniques to move beyond fixed target stimuli to precise neuromodulation.”
sauce:
South Carolina Medical College
Journal Reference:
Li, X. , et al. (2025). Salience Network Connectivity predicts response to repeated transcranial magnetic stimuli in smoking cessation: a preliminary machine learning study. Brain connections. doi.org/10.1177/21580014251376722
