Prediction of acute hypotensive episodes by means of neural network multi-models

Rocha, Teresa; Paredes, S.; Carvalho, Paulo; Henriques, J.

doi:10.1016/j.compbiomed.2011.07.006

Cited by 39 publications

(26 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, since similar blood pressure patterns also exist across both population groups, a lot of negative instances are incorrectly classified as hypotensive. Similar experiments on MIMIC-II by Rocha et al [50] demonstrated a sensitivity of 82.8% and a specificity of 78.4%. In another study, Lee and Mark [15] also demonstrated highest accuracies of 76% for single-mode and 86% for multi-mode datasets extracted from MIMIC-II.…”

Section: Prediction Resultssupporting

confidence: 65%

Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure

Ghosh

Feng

Nguyen

et al. 2016

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Acute hypotension is a significant risk factor for in-hospital mortality at intensive care units (ICUs). Prolonged hypotension can cause tissue hypoperfusion, leading to cellular dysfunction and severe injuries to multiple organs. Prompt medical interventions are thus extremely important for dealing with acute hypotensive episodes (AHE). Population level prognostic scoring systems for risk stratification of patients are suboptimal in such scenarios. However, the design of an efficient risk prediction system can significantly help in the identification of critical care patients, who are at risk of developing an AHE within a future time span. Towards this objective, a pattern mining algorithm is employed to extract informative sequential contrast patterns from hemodynamic data, for the prediction of hypotensive episodes. The hypotensive and normotensive patient groups are extracted from the MIMIC-II critical care research database, following an appropriate clinical inclusion criteria. The proposed method consists of a data preprocessing step to convert the blood pressure time series into symbolic sequences, using a symbolic aggregate approximation algorithm. Then, distinguishing subsequences are identified using the sequential contrast mining algorithm. These subsequences are used to predict the occurrence of an AHE in a future time window separated by a user-defined gap interval. Results indicate that the method performs well in terms of the prediction performance as well as in the generation of sequential patterns of clinical significance. Hence, the novelty of sequential patterns is in their usefulness as potential physiological biomarkers for building optimal patient risk stratification systems and for further clinical investigation of interesting patterns in critical care patients.

show abstract

Section: Prediction Resultssupporting

confidence: 65%

Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure

Ghosh

Feng

Nguyen

et al. 2016

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

show abstract

“…Lehman et al [20] carried out classification and forecasting tasks, using a similarity-based searching and pattern matching algorithm. More recently, Rocha et al [3] proposed neural network multi-models to predict adverse AHE occurring in ICU.…”

Section: Discissionmentioning

confidence: 99%

PhysioCover: Recovering the Missing Values in Physiological Data of Intensive Care Units

Kim¹,

Yang²,

Kim³

et al. 2014

International Journal of Contents

View full text Add to dashboard Cite

show abstract

“…Most of them used training and test datasets in the Physionet-Computers in Cardiology 2009 Challenge, except [7][8][9][10][11][12] in which they selected large dataset from MIMIC-II. In this paper each sample is related to one patient and the first experience of AHE is predicted, but in [7][8][9][10][11][12] some records were divided into some segments as a sample in which both groups (H and C) may contain one patient. RESP-SPO2 Table 3 depicts the ability of the proposed method before and after using hybrid features and feature selection in comparison with other methods.…”

Section: Acute Hypotensive Episode Prediction Resultsmentioning

confidence: 99%

“…In [7] generalized regression neural network models integrated in to a multimodel structure was proposed for prediction of MAP signal and diagnosis of AHE. Considering other vital signals in this methodology may reduce the number of multimodels or increased the horizon of prediction.…”

Section: Introductionmentioning

confidence: 99%

Predicting acute hypotensive episode by using hybrid features and a neuro-fuzzy network

Abbasinia¹,

Farokhi

Javadi³

2016

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Prediction of acute hypotensive episodes by means of neural network multi-models

Cited by 39 publications

References 11 publications

Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure

Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure

PhysioCover: Recovering the Missing Values in Physiological Data of Intensive Care Units

Predicting acute hypotensive episode by using hybrid features and a neuro-fuzzy network

Contact Info

Product

Resources

About