Wavelet based time series forecast with application to acute hypotensive episodes prediction

This paper presents a generic methodology for time series prediction, based on a wavelet decomposition/ reconstruction technique, together with a feedforward neural networks structure. The proposed methodology combines the flexibility and learning abilities of neural networks with a compact description of the signals, inherent to wavelets. In a first phase a wavelet decomposition of the signal is performed, providing a small number of coefficients that summarizes signal time evolution dynamics. The prediction problem is then effectively addressed by means of a neural networks model, previously trained using coefficients of the training dataset. The particular problem of forecasting acute hypotensive episodes (AHE) occurring in intensive care units was used to prove the effectiveness of the proposed strategy. The dataset, extracted from MIMIC-II, was made available in the context of the PhysioNet-Computers in Cardiology Challenge 2009. Results attained in this work were similar to the best ones achieved under that challenge.

[1]  Stéphane Canu,et al.  The long-term memory prediction by multiscale decomposition , 2000, Signal Process..

[2]  G.B. Moody,et al.  Similarity-based searching in multi-parameter time series databases , 2008, 2008 Computers in Cardiology.

[3]  Erol Cavus,et al.  Heart Rate Variability Predicts Severe Hypotension after Spinal Anesthesia , 2006, Anesthesiology.

[4]  Michel Camilleri.  Forecasting Using Non-Linear Techniques In Time Series Analysis : An Overview Of Techniques and Main Issues , 2004 .

[5]  Jules Bassale.  Hypotension Prediction Arterial Blood Pressure Variability , 2001 .

[6]  Stephen J. Roberts,et al.  Hierarchy, priors and wavelets: structure and signal modelling using ICA , 2004, Signal Process..

[7]  JH Henriques,et al.  Prediction of acute hypotensive episodes using neural network multi-models , 2009, 2009 36th Annual Computers in Cardiology Conference (CinC).

[8]  M. Frölich,et al.  Baseline heart rate may predict hypotension after spinal anesthesia in prehydrated obstetrical patients , 2002, Canadian journal of anaesthesia = Journal canadien d'anesthesie.

[9]  James McNames,et al.  Precursors In The Arterial Blood Pressure Signal To Episodes Of Acute Hypotension In Sepsis , 2002 .

[10]  G..  A Theory for Multiresolution Signal Decomposition : The Wavelet Representation , 2004 .

[11]  Gary Montague,et al.  Non-linear principal components analysis using genetic programming , 1997 .

[12]  Murali Tummala,et al.  Multirate, multiresolution, recursive Kalman filter , 2000, Signal Process..

[13]  James McNames,et al.  Local averaging optimization for chaotic time series prediction , 2002, Neurocomputing.

[14]  Olena Vynokurova,et al.  An adaptive learning algorithm for a wavelet neural network , 2005, Expert Syst. J. Knowl. Eng..

[15]  Pedro M. Domingos.  The Role of Occam's Razor in Knowledge Discovery , 1999, Data Mining and Knowledge Discovery.

[16]  Maciej Lawrynczuk,et al.  Computationally Efficient Nonlinear Predictive Control Based on RBF Neural Multi-models , 2009, International Conference on Adaptive and Natural Computing Algorithms.

[17]  Maciej Ławryńczuk,et al.  Computationally efficient nonlinear predictive control based on RBF neural multi-models , 2009, ICANNGA'09 2009.

[18]  Y. H. Song,et al.  Wavelet transform and neural networks for short-term electrical load forecasting , 2000 .

[19]  D. Reich,et al.  Predictors of Hypotension After Induction of General Anesthesia , 2005, Anesthesia and analgesia.

[20]  Rik Pintelon,et al.  Identification of Linear Systems: A Practical Guideline to Accurate Modeling , 1991 .

[21]  Zhong Shi-sheng,et al.  Time series prediction using wavelet process neural network , 2008 .

[22]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  Dominik R. Dersch,et al.  Multiresolution Forecasting for Futures Trading , 2001 .

[24]  M.E. El-Hawary,et al.  Short term load forecasting by using wavelet neural networks , 2000, 2000 Canadian Conference on Electrical and Computer Engineering. Conference Proceedings. Navigating to a New Era (Cat. No.00TH8492).

[25]  S. Cavalcanti,et al.  Short term variability of oxygen saturation during hemodialysis is a warning parameter for hypotension appearance , 2008, 2008 Computers in Cardiology.

[26]  G. Moody,et al.  Predicting acute hypotensive episodes: The 10th annual PhysioNet/Computers in Cardiology Challenge , 2010, 2009 36th Annual Computers in Cardiology Conference (CinC).