A wearable system enabling acute stress monitoring and closed-loop mitigation through transcutaneous median nerve...

Acute stress, in moderation, helps to prepare the body to overcome mental and physical challenges. However, excessive bouts of acute stress can be detrimental to the cardiovascular system and are a risk factor for cardiovascular disease and sudden cardiac death. Transcutaneous median nerve stimulation (tMNS) is a promising therapy for the mitigation of acute stress through peripheral neuromodulation, but the optimal delivery of tMNS for stress mitigation may require continuous monitoring of acute stress events for targeted delivery.

The purpose of this study was to develop a wearable system capable of continuous closed-loop acute stress monitoring and mitigation through non-invasive cardiovascular sensing and tMNS respectively.

A wearable wrist-worn device capable of sensing three channels of photoplethysmogram (PPG) and tri-axial accelerometry was designed. Pulse rate (PR) and PPG amplitude (PPGamp) were extracted from the acquired green PPG signal, while tMNS was delivered at varying intensities by custom-designed analog circuitry onboard the device. A companion app was used to wirelessly set stimulation levels by communicating with the device's microcontroller using Bluetooth low energy. The device was validated against bench-top sensors in a study with 19 healthy participants involving acute mental and physical stressors as well as tMNS. Repeated-measures correlation and Bland-Altman analyses were performed to compare PR extracted from 9904 5-s windows of PPG from the device and heart rate (HR) extracted beat-by-beat from bench-top electrocardiogram (ECG) and averaged across the same windows. Statistical tests were also performed to analyze differences in mean PR and PPGamp from baseline metrics across the acute stress and tMNS protocol.

PR extracted from our device correlated (r = 0.871, p < 0.001) and agreed (mean difference: 0.51 bpm, 95 % limits of agreement: 6.58 bpm, 5.57 bpm) strongly with HR extracted from bench-top ECG. We found decreases in mean PPGamp from baseline during stressors, while application of tMNS alongside stressors increased PPGamp back to baseline levels, and continued delivery of tMNS post-stressor further increased PPGamp to a significant difference from baseline. Significant reductions in PR as compared to baseline post-physical stressor also mirrored these findings, suggesting that our wearable device can track elevations in acute stress through cardiovascular monitoring while also mitigating the effects of acute stress through tMNS.

Our device is the first wearable, to our knowledge, to enable continuous monitoring of acute stress through cardiovascular sensing and feature extraction while mitigating acute stress through peripheral neuromodulation. Future work should test the device in ambulatory settings and investigate potential applications for clinical use-cases such as anxiety or trauma disorders.