Non-Invasive Multidimensional Capacitive Sensing for In Vivo Traumatic Brain Injury Monitoring

Traumatic brain injury (TBI) is a major cause of death and disability, but invasive intracranial pressure (ICP) monitoring is risky, and current non-invasive methods lack the resolution and reliability needed for continuous clinical use. We present a multidimensional, non-invasive single-electrode capacitance (SEC) sensing system for continuous TBI monitoring. Carbon-nanotube paper composite (CPC) electrodes detect regional ICP changes through permittivity variations driven by cerebrovascular pulsations, cerebrospinal fluid (CSF) thickness changes, and brain tissue microvibrations. Surrogate tissue model testing confirms sensitivity to CSF layer thickness, water layer height, vessel wall thickness, and vessel diameter. In vivo pig testing, we analyze the correlation between multisite SEC signals and ICP to identify potential novel sensing metrics. In TBI experiments, these metrics are further evaluated by examining dynamic, cross-hemispheric relationships using four SEC sensors to capture spatially resolved pathophysiology before and after injury. Statistical and machine-learning (ML) approaches are applied to derive novel digital markers and to estimate conventional indices of cerebral autoregulation and intracranial compliance from non-invasive features. Together, this wearable system enables portable, spatially resolved neurocritical monitoring across diverse clinical environments.