Forecasting intracranial hypertension using multi-scale waveform metrics

Abstract

Objective: Acute intracranial hypertension is an important risk factor of secondary brain damage after traumatic brain injury. Hypertensive episodes are often diagnosed reactively, leading to late detection and lost time for intervention planning. A pro-active approach that predicts critical events several hours ahead of time could assist in directing attention to patients at risk. Approach: We developed a prediction framework that forecasts onsets of acute intracranial hypertension in the next 8 hours. It jointly uses cerebral auto-regulation indices, spectral energies and morphological pulse metrics to describe the neurological state of the patient. One-minute base windows were compressed by computing signal metrics, and then stored in a multi-scale history, from which physiological features were derived. Main results: Our model predicted events up to 8 hours in advance with alarm recall rates of 90% at a precision of 30.3% in the MIMIC-III waveform database, improving upon two baselines from the literature. We found that features derived from high-frequency waveforms substantially improved the prediction performance over simple statistical summaries of low-frequency time series, and each of the three feature classes contributed to the performance gain. The inclusion of long-term history up to 8 hours was especially important. Significance: Our results highlight the importance of information contained in high-frequency waveforms in the neurological intensive care unit. They could motivate future studies on pre-hypertensive patterns and the design of new alarm algorithms for critical events in the injured brain.

Near-Infrared Spectroscopy to Assess Cerebral Autoregulation and Optimal Mean Arterial Pressure in Patients With Hypoxic-Ischemic Brain Injury: A Prospective Multicenter Feasibility Study

“…Comparing different lengths of the multi-window weighted algorithm (e.g., 2 vs 4 vs 8 hr) and the impact on MAP OPT would also be of interest, as this technique has been previously explored in patients with TBI ( 15 ) and delirium ( 36 ). Additionally, there may be other methods (e.g., differential entropy) to examine for higher-order relationships and improve the assessment of MAP OPT ( 37 ).…”

Non-Invasive Modeling of Intracranial Hypertension from Physiological Channels

“…Traumatic brain injury (TBI) is the main reason for the secondary human brain injury and death on the world. Although, the primary injury in (TBI) patients is diagnosed using brain imaging, the secondary brain injury, which may also lead to the following long term and/or permanent brain impairment remain as the main risk [1] such as: cerebral hypoxia defined as low brain oxygen flow, brain herniation described as brain swelling that may result in brain compression in skull and cerebral ischemia defined as low blood level in brain [2].…”

“…Prediction of intracranial hypertension (ICH) and patient outcome (survival, vegetarian life, death) based on the current health status and clinical recordings is crucially important for physicians [2]. Considering the important negative side effect and complexities associated with the invasive ICP monitoring various research groups have been looking into non-invasive approached [5][9].…”

Künstliche Intelligenz in der Neurointensivmedizin

“…Hier kommen die vorher erwähnten MIMIC-Datensätze zum Einsatz, um an Daten von 100 bis 120 Patienten mit einer Abtastrate von 50 Hz Vorhersagen zu tätigen. Es gelang mit bis zu einer AUC-ROC von 0,87, innerhalb von 6 h vor der hypertensiven Phase diese vorherzusagen [ 23 , 34 , 35 ].…”

“…Patienten, die auf der Neurointensivstation behandelt werden, leiden meist an akuten Erkrankungen. Hierzu zählen am häufigsten traumatische Hirnverletzungen, Blutungen (intrazerebrale, subdurale und Subarachnoidalblutung), maligne Schlaganfälle, epileptische Serien oder [23,34,35].…”