In neurointensive care, increased intracranial pressure (ICP) is a feared secondary brain insult in traumatic brain injury (TBI). A system that predicts ICP insults before they emerge may facilitate early optimization of the physiology, which may in turn lead to stopping the predicted ICP insult from occurring. The aim of this study was to evaluate the performance of different artificial intelligence models in predicting the risk of ICP insults. The models were trained to predict risk of ICP insults starting within 30 min, using the Uppsala high frequency TBI dataset. A restricted dataset consisting of only monitoring data were used, and an unrestricted dataset using monitoring data as well as clinical data, demographic data, and radiological evaluations was used. Four different model classes were compared: Gaussian process regression, logistic regression, random forest classifier, and Extreme Gradient Boosted decision trees (XGBoost). Six hundred and two patients with TBI were included (total monitoring 138,411 h). On the task of predicting upcoming ICP insults, the Gaussian process regression model performed similarly on the Uppsala high frequency TBI dataset (sensitivity 93.2

Critical care is complex and stressful. It is difficult to register in real time data not recorded by automatic systems. Time-specific knowledge of manual measures is important for understanding pathophysiology and for analyzing treatment and quality of care. Therefore, a novel iPad-based method for registration of manual measures was developed, which many can build themselves. Using a configuration for intracranial pressure (ICP) management, the methodology was validated, ICP treatment captured, and the quality of ICP management evaluated. Twenty-two patients with acute brain injuries were studied. The iPad-system was totally used for 2538 h. Thirteen-hundred-five manual measures were entered. Thirty-nine episodes of predefined ICP insults were identified. During 16/39 episodes, ICP treatments were registered. For 4/39 episodes treatments were registered within 90 s before or after the episode. For 3/39 episodes it was registered that treatment was intentionally refrained. In 15/16 episodes without registered treatment, the insult was mild or reasonable explanations were found when medical records and the Patient data management system were reviewed. In one situation without particular circumstances, morphine and clonidine were given to decrease ICP but not registered. No episodes of downtime or loss of data occurred. The developed methodology appears to be stable and robust as well as feasible and user-friendly. It was possible to capture the treatment of ICP insults with high temporal resolution, and to evaluate the quality of ICP management. An own developed novel tablet-based system like our system may be a promising potential tool useful in various future intensive care applications.

INTRODUCTION AND IMPORTANCE:Hypertensive crisis may be a life-threatening condition to any patient and represents an even more serious condition in trauma patients following severe hemorrhage.CASE PRESENTATION:We present a case were surgical drape packing induced hypertensive crisis in a trauma patient, recently resuscitated from abdominal hemorrhage.CLINICAL DISCUSSION:We argue that direct compression of the kidney by the surgical drapes induced hypersecretion of renin with a mechanism equal to Page kidney. The hypertensive crisis as well as the hyperreninemia was resolved after removing the surgical drapes, and the patient's condition returned to normal without any sequelae.CONCLUSION:We encourage considering this unusual but important complication when packing of the abdomen has been carried out, and strongly recommend ruling out renin-mediated hypertension as a cause of post-operative hypertension in such cases.

This paper describes a novel general method for automatic programming which can be seen as a generalization of techniques such as genetic programming and ADATE. The approach builds on the assumption that data compression can be used as a metaphor for cognition and intelligence. The proof-of-concept system is evaluated on sequence prediction problems. As a starting point, the process of inferring a general law from a data set is viewed as an attempt to compress the observed data. From an artificial intelligence point of view, compression is a useful way of measuring how deeply the observed data is understood. If the sequence contains redundancy it exists a shorter description i.e. the sequence can be compressed.