The goal of this observational study is to evaluate the diagnostic accuracy of non-invasive intracranial pressure (ICP) and intracranial compliance (ICC) pulse morphology and associated parameters (such as the P2/P1 ratio and nTTP) obtained with the brain4care system for the screening of intracranial hypertension (ICH) in patients with traumatic brain injury (TBI) and stroke treated in Brazil's public health system (SUS).

The main questions it aims to answer are:

Can non-invasive ICP and ICC pulse morphology reliably identify or exclude intracranial hypertension, cerebral edema, and hemorrhage compared to CT findings and clinical/neurological evaluations?

Can this approach differentiate ischemic stroke from hemorrhagic stroke with sufficient accuracy?

Does the use of brain4care contribute to earlier detection, improved clinical decision-making, and cost reduction in emergency settings?

This is a prospective, observational, multicenter, longitudinal cohort study conducted in four Brazilian states (Amazonas, São Paulo, Sergipe, and Rio Grande do Sul). The study aims to evaluate the diagnostic accuracy of non-invasive intracranial pressure (ICP) and intracranial compliance (ICC) monitoring, using the brain4care system, for screening intracranial hypertension (ICH) in patients with traumatic brain injury (TBI) and stroke (ischemic or hemorrhagic) admitted to multiple levels of care in the Brazilian Unified Health System (SUS), including pre-hospital care (SAMU), primary care (UBS), urgent care units (UPA), emergency rooms (PS), and intensive care units (ICUs).

Study Procedures

Eligible participants (≥18 years) with TBI or stroke will undergo daily non-invasive ICP and ICC monitoring using the BWS 2.0 (Braincare Wireless System). Monitoring will be performed by trained medical and nursing teams, adjusted according to care setting:

SAMU (pre-hospital): 5 minutes of valid pulses during stabilization and transport.

UBS, UPA, and PS: 10 minutes of valid pulses during initial evaluation; repeated daily until transfer.

ICU: three times daily (every 8 hours) for 10 minutes during the first 72h, then twice daily (every 12 hours) for up to 14 days, each session lasting 30 minutes.

Clinical data include demographics, comorbidities, neurological exam findings (Glasgow Coma Scale, Cincinnati Stroke Scale, NIHSS, Hunt-Hess, WFNS, Rankin), and vital signs. Neuroimaging will be collected when available, with application of validated CT-based scales (Marshall, Rotterdam, ASPECTS, Fisher, STICH, CRASH).

Technology

The BWS 2.0 sensor applies piezoelectric technology to detect minute cranial deformations linked to ICP waveforms, allowing calculation of morphological parameters such as P2/P1 ratio and normalized Time-to-Peak (nTTP). Changes in these parameters reflect alterations in intracranial compliance, providing a non-invasive tool to identify or exclude ICH.

Data Collection and Quality Assurance

All data will be recorded in REDCap, a secure, validated data capture system. A data dictionary describes all study variables, coding conventions (e.g., MedDRA for clinical events), and reference ranges. Automated range and consistency checks will be applied during data entry. Source data verification will be performed by comparing REDCap records with hospital charts, imaging reports, and device outputs.

Site monitoring will ensure adherence to standard operating procedures (SOPs) for patient recruitment, informed consent, device handling, data collection, and reporting of adverse events. Periodic audits and central data review will reinforce data integrity.

Sample Size and Analysis

An initial sample of 300 patients was calculated to provide sufficient power for ROC curve analysis. The primary outcome is diagnostic performance (AUC, sensitivity, specificity, predictive values) of brain4care parameters against reference standards (CT and clinical scales). Secondary analyses include correlation with clinical and imaging findings, differentiation of ischemic vs. hemorrhagic stroke, and evaluation of cases with hemorrhagic transformation.

Statistical methods include parametric or non-parametric tests (t-test, Mann-Whitney, ANOVA, Kruskal-Wallis), Pearson/Spearman correlations, and multivariate regression models. ROC analysis will quantify discriminatory capacity. Economic evaluation will assess cost-effectiveness, cost-benefit, and resource utilization, considering reduced need for invasive monitoring and imaging.

Risk and Benefit

Risks are minimal, limited to local skin irritation from sensor placement. Potential benefits include earlier recognition of ICH, reduced reliance on invasive monitoring, improved triage decisions, and optimized healthcare resource allocation in the SUS.

Inclusion Criteria

Patients of both sexes.

Age ≥ 18 years.

Clinical signs and symptoms of traumatic brain injury (mild, moderate, or severe) or stroke (ischemic or hemorrhagic).

Signed informed consent form by the patient or by a legally authorized representative in cases where the patient is unable to consent due to the severity of the clinical condition. If the legal representative is not present at the time of admission or initial care, consent may be obtained retrospectively.

Exclusion Criteria

Individuals with brain lesions of other etiologies, such as previous TBI and/or stroke, neoplasms, infections, sepsis, arteriovenous malformation, intoxication, or illicit drug use.

Individuals with scalp or skin lacerations that prevent proper placement of the brain4care sensor.

Individuals with craniectomy or skull fractures.

Individuals with severe debilitating mental health disorders and/or severe debilitating neurological diseases.

Individuals who are agitated, aggressive, or uncooperative, preventing follow-up and outcome assessment.

Pregnant women.

Adult patients (≥18 years) of both sexes presenting with clinical signs and symptoms of traumatic brain injury (mild, moderate, or severe) or stroke (ischemic or hemorrhagic). Participants will be recruited from multiple levels of care within the Brazilian Unified Health System (SUS), including primary care units (UBS), urgent care units (UPA), pre-hospital emergency services (SAMU), hospital emergency rooms (PS), and intensive care units (ICUs) at participating hospitals in the states of Amazonas, São Paulo, Sergipe, and Rio Grande do Sul.

Individual participant data will not be shared due to ethical and regulatory restrictions imposed by local ethics committees and data protection regulations.