Artefacts in continuous neuromonitoring

Abstract

Continuous neuromonitoring is essential in neurocritical care units, providing real-time insights into dynamic cerebral physiology for patients with neurological conditions, such as stroke and neurotrauma. Multiple modalities such as intracranial pressure, arterial blood pressure and near-infrared spectroscopy enable high-resolution dynamic tracking of cerebral perfusion pressure, and related variables critical for life-saving decisions. However, inherent, commonly unavoidable artefacts obscure insights into patient states and complicate treatment decisions. In this Review, we explore primary sources of artefacts in continuous neuromonitoring modalities, including clinical procedure activities, patient-related physiology, technical equipment properties and environmental factors, and their impacts on data integrity and clinical implications. We discuss emerging artefact management strategies, including domain knowledge and data-driven methods to mitigate impact and enhance data reliability. Additionally, we identify key translational challenges, indications for neurosensor design, harmonization and future artificial intelligence pathways, highlighting the need for robust, automated, real-time artefact management to enable precise, individualized patient care. Continuous neuromonitoring is vital for guiding neurocritical care, yet artefacts frequently obscure cerebral physiological signals and hinder clinical decisions. In this Review, sources of artefacts and their clinical impact are examined, alongside emerging strategies — both knowledge-based and data-driven — for artefact mitigation, sensor design and artificial intelligence-enabled real-time monitoring in precision neurocare.