Background: Traumatic brain injury (TBI) causes temporary or permanent alteration in brain functions. Generally, at intensive care units (ICU), intracranial pressure (ICP) is monitored and treated to avoid increases in ICP with associated secondary insults and poor clinical outcome. The aim of this study was to develop and evaluate a model which could predict future ICP levels of individual patients during their treatment in the ICU, and thus help the treating clinician to take proper actions before secondary injuries occure.
Methods: A simple, explainable, probabilistic Markov model was developed for the prediction task of ICP≥20 mmHg. Predictions were made for consecutive 10-minute intervals during the following hour, based on the preceding hour of ICP data. An easily implementable enhancement method was also developed and applied to compensate for imbalance in the data. The model was evaluated in a randomized and leave-one-out fashion on data from 29 patients with severe TBI.
Results: With random data selection from all patients (80/20% training/testing) and including the new enhancement method, the specificity of the model was high (93.9-95.0) and the sensitivity was good to high (72.7-87.1). Levels were similar (specificity 90.1-95.3 and sensitivity 73.0-88.7) when the model was trained by the leave-one-out method and evaluated on individual subjects.
Conclusion: The new model predicted increased levels of ICP in a reliable manner and the enhancement method to compensate for imbalanced data further improved the predictions. This lays the foundation for development of a bedside warning system designed to proactively avoid increased ICP levels in patients with severe TBI. Further advantages are the straightforward expandability of the model, enabling inclusion of other time series data and/or static parameters, making future studies of predictive strength based on combinations of physiological data possible. Next step is to evaluate the model on a larger patient material and to include other parameters apart from ICP.