Harnessing Big Data in Neurocritical Care in the Era of Precision Medicine

Purpose of review To survey the literature around "big data" and "artificial intelligence" use, its challenges, and examples in clinical and research domains within neurocritical care. Recent findings Innovations in digital data storage and integration technology have motivated the active development of data-driven tools for neurocritical care practice. There has been progress towards harmonization of data dictionaries within the field to boost generalizability of studies. Numerous collaborative groups have cultivated datasets that will enable further study of disease detection, prediction, and management of critically ill neurologic patients. Essential to the effective analysis of big data in neurocritical care are the increasing relationships between clinicians and data scientists. There are multiple challenges related to the efficient and ethical use of big data, including data complexity, database stewardship and governance, data quality, and safe implementation of data-driven conclusions. Continued efforts towards optimally harnessing data will be crucial in neurocritical care. Critically ill neurologic patients generate an abundant amount of data in the course of routine management. Clinical research is evolving from benefiting the "average patient" to aiming to deliver more precise treatment to the individual patient. In neurocritical care, this has manifested through big data-for triage decisions, enhancing workflow, event detection, and outcome prediction.