Wearable Neural Interfacing System for REM Sleep Restoration and Enhancement

Overview

This study evaluates a wearable system (NEUSleeP) that combines overnight EEG recording with transcranial focused ultrasound (tFUS) targeted to deep brain structures involved in REM sleep regulation (e.g., subthalamic nucleus). The primary objective is to assess safety and estimate effects on REM sleep quantity and architecture; secondary objectives include changes in stress-related measures. Healthy adults aged 18-50, with or without subclinical sleep or stress complaints, will complete two consecutive overnight recordings: Night 1 (baseline, no stimulation) and Night 2 (tFUS, EEG-guided and timed to REM). Participants will complete stress questionnaires. fMRI is conducted using two paradigms: in an imaging-validation subset, pre- and post-stimulation scans are acquired in the same MRI-FUS session; in the two-night cohorts, scans are acquired the morning before and the morning after the FUS night to assess BOLD responses. Outcomes include REM time, REM percentage, number of REM periods, REM latency, safety/tolerability, and exploratory neuroimaging and self-reported stress measures. Findings will inform the feasibility of a wearable EEG-tFUS approach to modulate REM sleep and stress adaptation.

This study evaluates the safety, feasibility, and preliminary signals of effect of NEUSleeP, a wearable neural interface for closed-loop modulation of REM sleep using transcranial focused ultrasound (tFUS) targeted to the subthalamic nucleus (STN). The system integrates a flexible ultrasound transducer with bioadhesive hydrogel EEG electrodes in a patch designed for repeated overnight use. REM sleep is implicated in emotional regulation, memory processes, and stress adaptation. Existing noninvasive approaches have focused largely on NREM modulation; this study examines a REM-focused, target-directed approach to determine whether tFUS delivered during sleep can alter REM architecture and related outcomes. The study comprises four phases: Phase 1 (device functionality) enrolls four healthy volunteers for repeated bench and on-body checks (EEG signal quality, contact impedance stability, usability) over four weeks, with comparisons to standard clinical electrodes. Phase 2 (STN stimulation and imaging validation) uses structural MRI and acoustic modeling to configure STN targeting for both a reference research system (BrainSonix Pulsar 1002) and NEUSleeP. In up to 20 healthy volunteers, functional MRI is acquired immediately before and immediately after the same MRI-FUS session to characterize BOLD responses in STN and stress-related networks (e.g., amygdala, insula), using identical imaging protocols across platforms. Phase 3 (REM modulation in healthy volunteers) enrolls 16 adults for two consecutive overnight recordings: Night 1 baseline (no stimulation) and Night 2 tFUS (closed-loop stimulation time-locked to REM using EEG). Primary sleep outcomes include REM time, REM percentage, number of REM periods, and REM latency; safety and tolerability are recorded throughout. Phase 4 (REM modulation in participants with non-clinical sleep disturbance) enrolls 12 adults with elevated sleep complaints and perceived stress for the same two-night protocol; exploratory outcomes include self-reported stress measures collected around the FUS night. Two fMRI paradigms are used across phases. In Phase 2, a same-session MRI-FUS visit includes pre-stimulation and post-stimulation fMRI acquisitions in the same session (healthy imaging-validation subset). In Phases 3 and 4, participants complete morning-before (pre-FUS night) and morning-after (post-FUS night) fMRI sessions to assess changes around the overnight REM-timed stimulation. Safety procedures include MRI screening where applicable, continuous adverse event capture, stop rules for stimulation, and post-visit follow-up. Ultrasound exposure is controlled within established diagnostic ultrasound limits (e.g., MI \<= 1.9 and derated time-averaged intensity within applicable FDA limits), with device-level acoustic verification prior to use. Data monitoring is performed by the study team with predefined criteria for pausing or discontinuation. Findings will inform the feasibility of a wearable EEG-tFUS approach for REM-related modulation and guide parameters for future controlled trials.