A beam of sound, tuned to a frequency that passes harmlessly through the skull, can now quiet one of the brain’s most deeply buried fear centers without a single incision. In a double-blind, sham-controlled trial published in Molecular Psychiatry in early 2025, a team led by neuroscientist Wynn Legon at the Fralin Biomedical Research Institute demonstrated that low-intensity transcranial focused ultrasound, known as tFUS, reduced activity in the left amygdala compared to a sham procedure, as measured by functional MRI. “We showed for the first time in a rigorous, sham-controlled design that we can reach the amygdala noninvasively and change its activity,” Legon said in a university summary of the findings. The result marks the first rigorous evidence that a completely noninvasive tool can modulate a structure sitting centimeters below the brain’s surface, one that plays a central role in PTSD, generalized anxiety, and other trauma-related disorders.
The finding arrives at a moment of real clinical need. Standard treatments for trauma disorders, including SSRIs and prolonged exposure therapy, fail roughly one-third to one-half of patients, according to estimates from the U.S. Department of Veterans Affairs. Transcranial magnetic stimulation, or TMS, has proven useful for depression but cannot penetrate deep enough to reach the amygdala. Deep brain stimulation can, but it requires neurosurgery and implanted electrodes. Focused ultrasound occupies a new middle ground: deep reach without an operating room.
What the strongest evidence shows
The Molecular Psychiatry trial used a within-subject design, meaning each participant received both active tFUS and a sham treatment in counterbalanced order, then underwent fMRI scanning after each. Because every person served as their own control, the design filters out individual differences in brain anatomy, baseline anxiety, and placebo susceptibility. Active stimulation produced a measurable drop in left amygdala BOLD signal, the blood-oxygen-level-dependent marker that fMRI uses to track neural activity, relative to sham.
The same research program included a three-week, 15-session open-label treatment phase for participants with mixed mood, anxiety, and trauma-related diagnoses. Short-term scores on standardized mood and anxiety scales improved, though the open-label design means expectation effects cannot be fully ruled out for those behavioral outcomes.
Separate mechanistic work, published in Frontiers in Neural Circuits, confirmed that tFUS can modulate deep structures including the amygdala and entorhinal cortex, with changes visible on arterial spin labeling perfusion MRI and BOLD functional connectivity maps. Researchers observed region-specific shifts in blood flow and connectivity that matched the intended targets, suggesting the ultrasound beam can be steered with millimeter-scale precision while leaving surrounding tissue largely unaffected. No adverse events were reported.
Safety data go unusually deep
One of the most compelling pieces of the safety case comes from epilepsy patients who were already scheduled for temporal lobe resections. Researchers delivered tFUS to part of the tissue before surgery, then sent both sonicated and unsonicated samples to pathologists for microscopic examination. The resulting study in Brain Stimulation, conducted under an FDA Investigational Device Exemption, found no meaningful differences in neurons, glial cells, or blood vessels between exposed and unexposed tissue. No necrosis, no inflammation, no micro-hemorrhage. It is rare in neuromodulation research to have this kind of direct histological confirmation in human tissue.
Connecting ultrasound to fear learning
A separate line of experiments, published in Science Advances, tested whether amygdala-targeted tFUS could alter the core psychological processes that malfunction in PTSD: threat learning and extinction. In standard fear-conditioning paradigms, healthy volunteers learn to associate a neutral cue with an unpleasant stimulus, then undergo extinction training where the cue appears without the aversive outcome. Participants who received active tFUS showed disrupted acquisition and extinction of conditioned fear, measured by skin conductance responses.
For clinicians, this link matters. It connects a noninvasive brain intervention directly to the specific psychological mechanisms, the stubborn persistence of learned fear, that drive intrusive memories, hypervigilance, and avoidance behavior in trauma disorders. Earlier noninvasive tools could not reach the circuits where those processes originate.
Engineering is keeping pace
Delivering ultrasound through the human skull with therapeutic precision is an engineering challenge, not just a clinical one. Each person’s skull varies in thickness, density, and curvature, all of which distort the ultrasound beam. A multi-element array system described in Nature Communications showed that deep brain targets, including thalamic nuclei, can be modulated with high spatial precision while the patient lies inside an fMRI scanner, without a neurosurgical frame. The platform integrates real-time imaging, skull-correction algorithms, and electronically steered beams, allowing clinicians to individualize targeting based on each person’s anatomy.
These engineering demonstrations are enabling science rather than proof of therapeutic benefit, but they address a practical prerequisite: any future clinical deployment will demand millimeter accuracy and reproducibility across diverse skull types.
What remains uncertain
No published trial has tracked patients for six to twelve months after tFUS treatment, so whether symptom relief persists is unknown. The open-label treatment phase lasted three weeks. Whether patients would need periodic maintenance sessions, similar to the booster protocols sometimes used in TMS for depression, is an open question.
Dosing parameters also lack consensus. A dose-escalation study of right-amygdala tFUS in healthy volunteers noted that intensities used for neuromodulation may exceed the FDA’s diagnostic ultrasound output limit of 720 mW/cm², and that imaging safety thresholds do not necessarily translate to brain stimulation applications. The FDA’s own guidance acknowledges that ultrasound is non-ionizing but flags potential for tissue heating and cavitation, recommending the ALARA principle: as low as reasonably achievable. An expert consensus framework from the International Transcranial Ultrasonic Stimulation Safety and Standards consortium, or ITRUSST, published in 2024, defined “non-significant risk” exposure levels with explicit thresholds for mechanical and thermal risk. But these benchmarks are advisory, not binding regulatory standards, and current studies vary widely in pulse duration, repetition frequency, and total energy delivered.
Head-to-head comparisons with existing treatments do not yet exist. No trial has pitted tFUS against SSRIs, cognitive behavioral therapy, or prolonged exposure therapy. Published participant pools have been small and limited in demographic diversity, raising questions about generalizability across ages, ethnicities, and symptom severity. Many enrollees were already taking psychiatric medications, further complicating interpretation.
What larger trials must still prove about focused ultrasound for trauma disorders
As of June 2026, the evidence supports tFUS as a promising but experimental tool for modulating deep fear circuits. The strongest data concern target engagement and safety. Clinical efficacy, the question of whether patients actually feel and function better over the long term, is still being mapped out.
Regulators and clinicians will likely demand larger, longer, and more demographically diverse randomized trials before integrating tFUS into routine psychiatric care. The gap between reducing a conditioned skin-conductance response in a laboratory and relieving the intrusive memories and hypervigilance that define daily life with PTSD remains substantial. But the gap is narrower than it was even two years ago. If upcoming trials confirm durable symptom relief with acceptable risk, focused ultrasound could become a genuinely new option for the large share of trauma patients whose symptoms resist everything currently available.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
