Back

An updated prospective quantitative analysis of symptoms and safety in low-intensity focused ultrasound neuromodulation

Kapoor, A.; Crahan, T.; Legon, W.

2026-06-29 psychiatry and clinical psychology
10.64898/2026.06.25.26356569 medRxiv
Show abstract

Low-intensity focused ultrasound (LIFU) is a non-invasive neuromodulation technique with a favorable safety profile in healthy volunteers. Participant-experienced symptoms however remain inconsistently measured, and prospective benchmarks are lacking. Here, we prospectively characterized symptoms associated with LIFU neuromodulation across eight studies using a standardized Report of Symptoms (ROS). We compiled 629 sessions (472 LIFU, 157 sham) in 106 healthy adults (28.1 +/- 9.8 years) across eight cortical and subcortical targets (500 kHz; extracranial ISPPA 3.9-33.3 W/cm2; mechanical index 0.5-1.4). The ROS rated 17 symptom domains from 0 (absent) to 3 (severe) before and after each session. New-onset incidence, symptom severity, and total symptom burden were compared between LIFU and sham. The same instrument was applied in 35 patients with chronic pain. Symptom profiles after LIFU were indistinguishable from sham across all 17 domains. Total symptom burden averaged approximately one domain per session and did not increase after LIFU (0.94 to 1.03; p = 0.120). Post-intervention burden was predicted by baseline burden (beta = 0.347, p < 0.001) but not by stimulation condition (p = 0.222). New-onset symptoms did not increase across up to 27 LIFU sessions (OR = 0.99, p = 0.73) and were weakly, non-significantly related to acoustic intensity (rho = 0.37). Across a prospective, sham-controlled dataset, LIFU added no measurable symptom burden and was well tolerated in healthy adults, with comparable tolerability in patients. These findings establish a benchmark for the safety of human LIFU neuromodulation and a foundation for its therapeutic translation.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.