Wearable-Machine Interface Architectures

Wrist-worn wearable devices provide rich sets of pulsatile physiological data under various modalities and circumstances. An unexploited capability is that the pulsatile physiological time series collected by wrist-worn wearable devices can be used for recovering internal brain dynamics. This Cyber Physical System research project integrates computational algorithms for state-space estimation with wrist-worn wearable devices that sense the physiological signals to present wearable machine-interface architectures related to mental stress. One of  the goals of this projects is to enable inference of underlying neural mechanisms and reconstruction of mental-stress-related brain dynamics through wrist-worn wearable devices. In particular, we investigate inferring the underlying autonomic nervous system stimulation and decoding neurocognitive stress using galvanic skin response measurements.

The human body responds to neurocognitive stress in multiple ways through its autonomic nervous system. Changes in skin conductance measurements indicate sudomotor nerve activity, and could be used in inferring the underlying autonomic nervous system stimulation. We model skin conductance measurements using a state-space model with sparse impulsive events as inputs to the model. Then, we recover the timings and amplitudes of sudomotor nerve activity using a generalized cross-validation based sparse recovery approach. Finally, we relate stress to the probability that a phasic driver impulse occurs in skin conductance signals to continuously track a stress level. Results demonstrate a promising approach for tracking stress through wearable devices. Decoding brain states using wrist-worn wearables will transform how mental-stress-related diseases are diagnosed and treated.