Key result
Artificial neural networks trained on hemodynamic data predicted impending hypotensive episodes with a mean area under the ROC curve of 0.918, sensitivity of 0.826, and specificity of 0.859.
Why the study?
Can artificial neural networks trained on hemodynamic data predict impending hypotensive episodes in intensive care unit patients?
Observational (n=1,311)
Can artificial neural networks trained on hemodynamic data predict impending hypotensive episodes in intensive care unit patients?
Effect estimate: AUC 0.918
Artificial neural networks can identify discriminatory patterns in hemodynamic data to predict impending hypotensive episodes in the ICU, though low prevalence limits positive predictive value.
Supports ANN-based early warning for ICU hypotension; hypothesis-generating pending prospective outcome validation.
Background In the intensive care unit (ICU), clinical staff must stay vigilant to promptly detect and treat hypotensive episodes (HEs). Given the stressful context of busy ICUs, an automated hypotensive risk stratifier can help ICU clinicians focus care and resources by prospectively identifying patients at increased risk of impending HEs. The objective of this study was to investigate the possible existence of discriminatory patterns in hemodynamic data that can be indicative of future hypotensive risk. Methods Given the complexity and heterogeneity of ICU data, a machine learning approach was used in this study. Time series of minute-by-minute measures of mean arterial blood pressure, heart rate, pulse pressure, and relative cardiac output from 1,311 records from the MIMIC II Database were used. An HE was defined as a 30-minute period during which the mean arterial pressure was below 60 mmHg for at least 90% of the time. Features extracted from the hemodynamic data during an observation period of either 30 or 60 minutes were analyzed to predict the occurrence of HEs 1 or 2 hours into the future. Artificial neural networks (ANNs) were trained for binary classification (normotensive vs. hypotensive) and regression (estimation of future mean blood pressure). Results The ANNs were successfully trained to discriminate patterns in the multidimensional hemodynamic data that were predictive of future HEs. The best overall binary classification performance resulted in a mean area under ROC curve of 0.918, a sensitivity of 0.826, and a specificity of 0.859. Predicting further into the future resulted in poorer performance, whereas observation duration minimally affected performance. The low prevalence of HEs led to poor positive predictive values. In regression, the best mean absolute error was 9.67%. Conclusions The promising pattern recognition performance demonstrates the existence of discriminatory patterns in hemodynamic data that can indicate impending hypotension. The poor PPVs discourage a direct HE predictor, but a hypotensive risk stratifier based on the pattern recognition algorithms of this study would be of significant clinical value in busy ICU environments.
No takes yet. Share an insight, caveat, or question.
Roger G. Mark (2010) conducted an observational in Impending hypotension in intensive care (n=1,311). Artificial neural network pattern recognition vs. Normotensive periods was evaluated on Prediction of hypotensive episodes (binary classification) (AUC 0.918). Artificial neural networks trained on hemodynamic data predicted impending hypotensive episodes with a mean area under the ROC curve of 0.918, sensitivity of 0.826, and specificity of 0.859.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: