Researchers at Northwestern University have introduced a novel framework for personalizing intermittent theta burst stimulation using real-time electroencephalography feedback. The approach, detailed in a new preprint, aims to synchronize stimulation trains with individual prefrontal theta rhythms to potentially enhance precision in neuromodulation techniques.

Background on Transcranial Magnetic Stimulation and Theta Burst Protocols

Transcranial magnetic stimulation, commonly abbreviated as TMS, delivers magnetic pulses to targeted brain regions to modulate neural activity without invasive procedures. Among TMS variants, intermittent theta burst stimulation, or iTBS, has gained attention for its efficiency in inducing lasting changes in cortical excitability. Standard iTBS protocols deliver bursts of pulses at theta frequencies, typically mimicking natural brain rhythms around 5 Hz, over short trains separated by longer intervals.

Traditional iTBS has shown promise in treating conditions such as depression, but response rates vary significantly across individuals. This variability has driven interest in closed-loop or state-dependent approaches that adapt stimulation to ongoing brain activity measured via electroencephalography, known as EEG.

The Challenge of EEG-Synchronized iTBS

Aligning iTBS with real-time EEG signals presents unique technical hurdles. Stimulation pulses within each train interfere with EEG recordings, making continuous phase tracking difficult after the initial pulse. Prefrontal theta oscillations, targeted in many psychiatric applications, also tend to be less stable and lower in amplitude compared to motor cortex rhythms often studied in earlier closed-loop TMS work.

Previous studies have demonstrated that the phase of ongoing EEG oscillations can influence how susceptible the cortex is to a single TMS pulse. Extending this principle to repetitive protocols like iTBS requires overcoming forecasting limitations and pulse artifacts.

Introducing the Personalized EEG-Controlled iTBS Framework

The new method, termed iTBS-EEG, calibrates parameters on each participant's individual EEG data before stimulation begins. It forecasts the optimal timing for initiating each stimulation train based on the person's dominant theta phase in the prefrontal cortex. Subsequent bursts within the train follow a personalized inter-burst interval derived from the same calibration.

Key to the approach is a "seed-and-sustain" hypothesis. By precisely timing the first few bursts to the endogenous theta phase, the protocol may entrain local oscillations, allowing later bursts to ride the sustained rhythm even if perfect multi-second forecasting proves impossible.

The system architecture includes high-dynamic-range EEG hardware for real-time streaming, dedicated computing units for monitoring and control, neuronavigation, and a collaborative robot for precise coil positioning. Total system delay from EEG sampling to coil discharge measures approximately 1.7 to 1.9 milliseconds.

Photo by Francesco Ungaro on Unsplash

Validation Experiments and Technical Feasibility

Two experiments assessed the framework's performance. The first used a large public dataset from the Dortmund Vital Study, involving over 600 participants aged 20 to 70. Calibration on one resting-state recording allowed accurate prediction of theta phase for train initiation in a second recording taken two hours later, with median accuracies exceeding 95 percent. Alignment for the second burst in each train reached around 70 percent.

A smaller in-situ validation at Northwestern University with 10 younger adults confirmed similar performance when equipment operated in real time, though actual pulses were withheld to preserve clean EEG signals. Parameters remained stable across typical session durations, supporting practical applicability.

These results indicate reliable train initiation across age groups and sexes, addressing concerns about generalizability in diverse populations.

Authors and Institutional Context at Northwestern University

The work is led by researchers affiliated with the Stephen M. Stahl Center for Psychiatric Neuroscience in the Department of Psychiatry and Behavioral Sciences at Northwestern University's Feinberg School of Medicine. Corresponding author Ivan Alekseichuk contributed conceptualization, methodology, and supervision, with co-authors Felix A. Maldonado Osorio, Mina Elhamiasl, Mary K. Gacek, and Stewart A. Shankman providing expertise in software development, validation, investigation, and project oversight.

Northwestern's interdisciplinary environment, combining psychiatry, psychology, and engineering, facilitated the technical innovations. Funding came from the National Institute of Mental Health under grant MH128454.

Read the full preprint at https://www.sciencedirect.com/science/article/pii/S1935861X26001208 or the medRxiv version for detailed methods and supplementary materials.

Implications for Precision Psychiatry and Neuromodulation Research

This technical demonstration opens pathways for more individualized brain stimulation protocols. By matching stimulation timing to a patient's unique theta rhythm, clinicians and researchers may one day reduce the trial-and-error element currently common in TMS therapies.

Potential applications extend beyond depression to other conditions involving prefrontal dysfunction, such as anxiety disorders or cognitive impairments. The seed-and-sustain concept also invites further mechanistic studies using combined TMS-EEG or intracranial recordings to verify entrainment effects.

University laboratories equipped with TMS-EEG setups stand to benefit from adopting similar calibration pipelines, fostering collaborative projects across neuroscience and engineering departments.

Future Directions and Clinical Translation Needs

The authors emphasize that clinical trials remain essential to determine whether EEG-controlled iTBS yields superior therapeutic outcomes compared to standard protocols. Questions around optimal phase targeting, session frequency, and integration with behavioral therapies require systematic investigation.

Broader adoption could influence training programs in academic medical centers, emphasizing skills in real-time signal processing, personalized medicine approaches, and regulatory considerations for closed-loop devices.

Ethical aspects, including informed consent for adaptive stimulation and data privacy in EEG recordings, will also shape implementation strategies.

Photo by Brett Jordan on Unsplash

Opportunities for Academic Researchers and Career Pathways

Publications like this highlight growing demand for expertise at the intersection of neuroscience, engineering, and clinical research. Graduate students and postdoctoral fellows pursuing work in neuromodulation may find expanded opportunities in university labs focused on precision brain stimulation.

Interdisciplinary programs combining psychology, biomedical engineering, and psychiatry prepare the next generation of investigators. Resources on academic career development, including positions in research-intensive institutions, support professionals seeking roles in these emerging fields.

Broader Context in Brain Stimulation Research Landscape

Closed-loop TMS approaches have advanced rapidly, with multiple groups exploring phase-dependent effects. This contribution from Northwestern adds a practical solution tailored to iTBS's repetitive structure, complementing single-pulse or open-loop personalization strategies reported elsewhere.

Continued refinement of forecasting algorithms and hardware integration promises to accelerate translation from bench to bedside in academic and clinical settings worldwide.