Autoregressive Hidden Markov Models for the Early Detection of Neonatal Sepsis

Late onset neonatal sepsis is one of the major clinical concerns when premature babies receive intensive care. Current practice relies on slow laboratory testing of blood cultures for diagnosis. A valuable research question is whether sepsis can be reliably detected before the blood sample is taken. This paper investigates the extent to which physiological events observed in the patient's monitoring traces could be used for the early detection of neonatal sepsis. We model the distribution of these events with an autoregressive hidden Markov model (AR-HMM). Both learning and inference carefully use domain knowledge to extract the baby's true physiology from the monitoring data. Our model can produce real-time predictions about the onset of the infection and also handles missing data. We evaluate the effectiveness of the AR-HMM for sepsis detection on a dataset collected from the Neonatal Intensive Care Unit at the Royal Infirmary of Edinburgh.

[1]  Lalit R. Bahl,et al.  Maximum mutual information estimation of hidden Markov model parameters for speech recognition , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Vijay V. Raghavan,et al.  A critical investigation of recall and precision as measures of retrieval system performance , 1989, TOIS.

[3]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[4]  Biing-Hwang Juang,et al.  On the application of hidden Markov models for enhancing noisy speech , 1989, IEEE Trans. Acoust. Speech Signal Process..

[5]  G.G. Cano,et al.  An approach to cardiac arrhythmia analysis using hidden Markov models , 1990, IEEE Transactions on Biomedical Engineering.

[6]  R. Shumway,et al.  Dynamic linear models with switching , 1991 .

[7]  Philip C. Woodland,et al.  Hidden Markov models using vector linear prediction and discriminative output distributions , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  D. Haussler,et al.  Hidden Markov models in computational biology. Applications to protein modeling. , 1993, Journal of molecular biology.

[9]  R. Ehrenkranz,et al.  Bloodstream infections in neonatal intensive care unit patients: results of a multicenter study. , 1994, The Pediatric infectious disease journal.

[10]  Anders Krogh,et al.  Hidden Neural Networks , 1999, Neural Computation.

[11]  M. P. Griffin,et al.  Toward the early diagnosis of neonatal sepsis and sepsis-like illness using novel heart rate analysis. , 2001, Pediatrics.

[12]  W. Poole,et al.  Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. , 2002, Pediatrics.

[13]  Hinrich Schütze,et al.  Book Reviews: Foundations of Statistical Natural Language Processing , 1999, CL.

[14]  F. Harrell,et al.  Abnormal Heart Rate Characteristics Preceding Neonatal Sepsis and Sepsis-Like Illness , 2003, Pediatric Research.

[15]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[16]  Tom Fawcett,et al.  Robust Classification for Imprecise Environments , 2000, Machine Learning.

[17]  Michael I. Jordan,et al.  Factorial Hidden Markov Models , 1995, Machine Learning.

[18]  M. T. Johnson,et al.  Capacity and complexity of HMM duration modeling techniques , 2005, IEEE Signal Processing Letters.

[19]  Richard Simon,et al.  Bias in error estimation when using cross-validation for model selection , 2006, BMC Bioinformatics.

[20]  Jianguo Zhang,et al.  The PASCAL Visual Object Classes Challenge , 2006 .

[21]  Steve Young,et al.  The HTK book version 3.4 , 2006 .

[22]  Douglas E. Lake,et al.  Heart rate characteristics monitoring for neonatal sepsis , 2006, IEEE Transactions on Biomedical Engineering.

[23]  David Barber,et al.  A Novel Gaussian Sum Smoother for Approximate Inference in Switching Linear Dynamical Systems , 2006, NIPS.

[24]  Bernadette Dorizzi,et al.  ECG signal analysis through hidden Markov models , 2006, IEEE Transactions on Biomedical Engineering.

[25]  M. Stacey,et al.  An architecture for multi-dimensional temporal abstraction and its application to support neonatal intensive care , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[26]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[27]  J. Quinn.  Bayesian Condition Monitoring in Neonatal Intensive Care , 2007 .

[28]  C. Doré,et al.  A case definition for national and international neonatal bloodstream infection surveillance , 2008, Archives of Disease in Childhood Fetal and Neonatal Edition.

[29]  John A. Quinn,et al.  Factorial Switching Linear Dynamical Systems Applied to Physiological Condition Monitoring , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  J. M. Eklund,et al.  Real-Time Analysis for Intensive Care: Development and Deployment of the Artemis Analytic System , 2010, IEEE Engineering in Medicine and Biology Magazine.

[31]  J. R. Moorman,et al.  Periodic heart rate decelerations in premature infants , 2010, Experimental biology and medicine.

[32]  W. Carlo,et al.  Mortality reduction by heart rate characteristic monitoring in very low birth weight neonates: a randomized trial. , 2011, The Journal of pediatrics.

[33]  Carolyn McGregor,et al.  Variability analysis with analytics applied to physiological data streams from the neonatal intensive care unit , 2012, 2012 25th IEEE International Symposium on Computer-Based Medical Systems (CBMS).

[34]  Ioan Stanculescu.  Dynamical models for neonatal intensive care monitoring , 2015 .