Knowledge-based approach to septic shock patient data using a neural network with trapezoidal activation functions

In this contribution we present an application of a knowledge-based neural network technique in the domain of medical research. We consider the crucial problem of intensive care patients developing a septic shock during their stay at the intensive care unit. Septic shock is of prime importance in intensive care medicine due to its high mortality rate. Our analysis of the patient data is embedded in a medical data analysis cycle, including preprocessing, classification, rule generation and interpretation. For classification and rule generation we chose an improved architecture based on a growing trapezoidal basis function network for our metric variables. Our results extend those of a black box classification and give a deeper insight in our patient data. We evaluate our results with classification and rule performance measures. For feature selection we introduce a new importance measure.

[1]  A. Fein,et al.  Sepsis and multiorgan failure , 1997 .

[2]  Michael R. Berthold,et al.  Building precise classifiers with automatic rule extraction , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[3]  Alan J. Miller,et al.  Subset Selection in Regression , 1991 .

[4]  E. Hanisch,et al.  Intensive Care Management in Abdominal Surgical Patients with Septic Complications , 2001 .

[5]  E. Faist.  Immunological Screening and Immunotherapy in Critically ill Patients with Abdominal Infections , 2001, Springer Berlin Heidelberg.

[6]  W. Baxt.  Application of artificial neural networks to clinical medicine , 1995, The Lancet.

[7]  Jürgen Paetz.  Metric rule generation with septic shock patient data , 2001, Proceedings 2001 IEEE International Conference on Data Mining.

[8]  Chuen-Tsai Sun,et al.  Functional equivalence between radial basis function networks and fuzzy inference systems , 1993, IEEE Trans. Neural Networks.

[9]  Shusaku Tsumoto,et al.  Clinical Knowledge Discovery in Hospital Information Systems: Two Case Studies , 2000, PKDD.

[10]  M. Musen,et al.  Handbook of Medical Informatics , 2002 .

[11]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[12]  Michael R. Berthold,et al.  Boosting the Performance of RBF Networks with Dynamic Decay Adjustment , 1994, NIPS.

[13]  Alfred Ultsch,et al.  Integration of Neural Networks and Knowledge-Based Systems in Medicine , 1995, AIME.

[14]  E. Hanisch,et al.  Epidemiologie von SIRS, Sepsis und septischem Schock bei chirurgischen Intensivpatienten , 1998, Der Chirurg.

[15]  Thomas Villmann.  Neural networks approaches in medicine - a review of actual developments , 2000, ESANN.

[16]  E. Neugebauer,et al.  Thirty years of anti-mediator treatment in sepsis and septic shock – what have we learned? , 1998, Langenbeck's Archives of Surgery.

[17]  Michael R. Berthold,et al.  From radial to rectangular basis functions : A new approach for rule learning from large datasets , 1995 .

[18]  Rüdiger W. Brause,et al.  A neuro-fuzzy approach as medical diagnostic interface , 2000, ESANN.

[19]  W Penny,et al.  Neural Networks in Clinical Medicine , 1996, Medical decision making : an international journal of the Society for Medical Decision Making.

[20]  Rudolf Kruse,et al.  Obtaining interpretable fuzzy classification rules from medical data , 1999, Artif. Intell. Medicine.

[21]  Paulo J. G. Lisboa,et al.  Artificial Neural Networks in Biomedicine , 2000, Perspectives in Neural Computing.

[22]  Detlef Nauck,et al.  Foundations Of Neuro-Fuzzy Systems , 1997 .

[23]  Jürgen Paetz,et al.  About the Analysis of Septic Shock Patient Data , 2000, ISMDA.

[24]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[25]  Johannes Ruhland,et al.  Enhancing Rule Interestingness for Neuro-fuzzy Systems , 1999, PKDD.

[26]  J. Nazuno.  Haykin, Simon. Neural networks: A comprehensive foundation, Prentice Hall, Inc. Segunda Edición, 1999 , 2000 .

[27]  I. Jolliffe.  Principal Component Analysis , 2002 .

[28]  Michael R. Berthold,et al.  Discriminative Power of Input Features in a Fuzzy Model , 1999, IDA.

[29]  J. Vincent,et al.  The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure , 1996, Intensive Care Medicine.

[30]  Ramakrishnan Srikant,et al.  Fast Algorithms for Mining Association Rules in Large Databases , 1994, VLDB.

[31]  Bernd Fritzke,et al.  Incremental neuro-fuzzy systems , 1997, Optics & Photonics.

[32]  Michael R. Berthold,et al.  Constructive training of probabilistic neural networks , 1998, Neurocomputing.

[33]  Jihoon Yang,et al.  Feature Subset Selection Using a Genetic Algorithm , 1998, IEEE Intell. Syst..

[34]  Hardaway Rm,et al.  A review of septic shock. , 2000 .

[35]  Martti Juhola,et al.  Treatment of missing data values in a neural network based decision support system for acute abdominal pain , 1998, Artif. Intell. Medicine.

[36]  Rüdiger W. Brause,et al.  A Frequent Patterns Tree Approach for Rule Generation with Categorical Septic Shock Patient Data , 2001, ISMDA.

[37]  Hiroshi Tsukimoto,et al.  Extracting rules from trained neural networks , 2000, IEEE Trans. Neural Networks Learn. Syst..

[38]  Michael R. Berthold.  Fuzzy models and potential outliers , 1999, 18th International Conference of the North American Fuzzy Information Processing Society - NAFIPS (Cat. No.99TH8397).

[39]  Ramakrishnan Srikant,et al.  Fast algorithms for mining association rules , 1998, VLDB 1998.

[40]  W. Vach,et al.  Neural networks and logistic regression: Part I , 1996 .

[41]  Jürgen Paetz.  Some remarks on choosing a method for outcome prediction. , 2002, Critical care medicine.

[42]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..