Machine Learning Applications in Personalized Transcranial Magnetic Stimulation (TMS) Therapy

Xiangbo Zhang

Co-Presenters: Individual Presentation

College: The Dorothy and George Hennings College of Science, Mathematics and Technology

Major: Computer Science

Faculty Research Mentor: Yousef Nejatbakhsh

Abstract:

Transcranial Magnetic Stimulation (TMS) is a non-invasive neuromodulationtechnique used to treat neurological and psychiatric disorders. However, optimizingTMS treatment remains challenging due to patient response variability. Machinelearning offers a powerful approach to enhancing TMS efficacy by leveraging brainimaging (MRI, fMRI) and electroencephalography (EEG) data to personalizetreatment parameters.This study explores key applications of machine learning in TMS,including predicting treatment response by identifying neurophysiologicalbiomarkers, optimizing coil placement for precise stimulation of target brain regions,and real-time adjustment of stimulation parameters based on EEG feedback.Additionally, machine learning aids in understanding the mechanisms of action byanalyzing large-scale datasets to reveal neural pathways responsive to TMS.Integrating machine learning with TMS allows treatment plans to be personalized toindividual brain characteristics, potentially improving clinical outcomes. Furthermore,machine learning-driven insights can contribute to advancing TMS research and developing more effective neuromodulation strategies. This research highlights theintersection of artificial intelligence and neurostimulation, demonstrating the potentialof data-driven approaches to refine and optimize TMS therapy.

Previous
Previous

Adversarial Testing of LLMs in Less Common Languages to Test and Analyze Vulnerabilities in ChatGPT and Gemini

Next
Next

Healthcare Utilization in Graduate Students vs Employed Adults