Comparing models for quantitative risk assessment: an application to the European Registry of foreign body injuries in children

Risk Assessment is the systematic study of decisions subject to uncertain consequences. An increasing interest has been focused on modeling techniques like Bayesian Networks since their capability of (1) combining in the probabilistic framework different type of evidence including both expert judgments and objective data; (2) overturning previous beliefs in the light of the new information being received and (3) making predictions even with incomplete data. In this work, we proposed a comparison among Bayesian Networks and other classical Quantitative Risk Assessment techniques such as Neural Networks, Classification Trees, Random Forests and Logistic Regression models. Hybrid approaches, combining both Classification Trees and Bayesian Networks, were also considered. Among Bayesian Networks, a clear distinction between purely data-driven approach and combination of expert knowledge with objective data is made. The aim of this paper consists in evaluating among this models which best can be applied, in the framework of Quantitative Risk Assessment, to assess the safety of children who are exposed to the risk of inhalation/insertion/aspiration of consumer products. The issue of preventing injuries in children is of paramount importance, in particular where product design is involved: quantifying the risk associated to product characteristics can be of great usefulness in addressing the product safety design regulation. Data of the European Registry of Foreign Bodies Injuries formed the starting evidence for risk assessment. Results showed that Bayesian Networks appeared to have both the ease of interpretability and accuracy in making prediction, even if simpler models like logistic regression still performed well.

[1]  Averill Law,et al.  Simulation Modeling and Analysis (McGraw-Hill Series in Industrial Engineering and Management) , 2006 .

[2]  Frederick P. Rivara,et al.  The Epidemiology and Causes of Childhood Injuries. , 1987 .

[3]  David Parkinson,et al.  Risk: Analysis, perception and management. report of a Royal Society Study Group: Pp 201. The Royal Society. 1992. Paperback £15.50 ISBN 0 85403 467 6 , 1993 .

[4]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[5]  William N. Venables,et al.  Modern Applied Statistics with S , 2010 .

[6]  Christina Gloeckner,et al.  Modern Applied Statistics With S , 2003 .

[7]  G Rider,et al.  Characteristics of objects that cause choking in children. , 1995, JAMA.

[8]  Yoshua Bengio,et al.  Pattern Recognition and Neural Networks , 1995 .

[9]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[10]  Louis Anthony Cox Detecting Causal Nonlinear Exposure-Response Relations in Epidemiological Data , 2006, Dose-response : a publication of International Hormesis Society.

[11]  D. Vose,et al.  The application of quantitative risk assessment to microbial food safety. , 1998, Journal of food protection.

[12]  Louis Anthony Cox,et al.  Risk Analysis Foundations, Models, and Methods , 2001 .

[13]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[14]  Norman E. Fenton,et al.  Software Measurement: Uncertainty and Causal Modeling , 2002, IEEE Softw..

[15]  Hugh A Barton,et al.  Framework for Evaluation of Physiologically‐Based Pharmacokinetic Models for Use in Safety or Risk Assessment , 2004, Risk analysis : an official publication of the Society for Risk Analysis.

[16]  David Heckerman,et al.  A Tutorial on Learning with Bayesian Networks , 1998, Learning in Graphical Models.

[17]  G Rider,et al.  Human factors project: development of computer models of anatomy as an aid to risk management. , 1998, International journal of pediatric otorhinolaryngology.

[18]  J. Wreathall,et al.  Assessing risk: the role of probabilistic risk assessment (PRA) in patient safety improvement , 2004, Quality and Safety in Health Care.

[19]  Chitra Rajagopal,et al.  Quantitative risk assessment for accidental release of titanium tetrachloride in a titanium sponge production plant. , 2003, Journal of hazardous materials.

[20]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[21]  Oliver Floerl,et al.  A Risk-Based Predictive Tool to Prevent Accidental Introductions of Nonindigenous Marine Species , 2005, Environmental management.

[22]  E. Steyerberg,et al.  [Regression modeling strategies]. , 2011, Revista espanola de cardiologia.

[23]  Norman E. Fenton,et al.  1 2 3 4 5 6 7 , 2001 .

[24]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[25]  J. Pearl Causality: Models, Reasoning and Inference , 2000 .

[26]  Michael P. Wellman,et al.  Real-world applications of Bayesian networks , 1995, CACM.

[27]  Terje Aven,et al.  Perspectives on risk: review and discussion of the basis for establishing a unified and holistic approach , 2005, Reliab. Eng. Syst. Saf..

[28]  Saul Rockman In school or out: technology, equity, and the future of our kids , 1995, CACM.

[29]  G. H. Kinchin Quantitative risk analysis , 1979 .

[30]  J L Pater,et al.  Comparing the predictive value of neural network models to logistic regression models on the risk of death for small-cell lung cancer patients. , 2006, European journal of cancer care.

[31]  John S Evans,et al.  A Regression‐Based Approach for Estimating Primary and Secondary Particulate Matter Intake Fractions , 2002, Risk analysis : an official publication of the Society for Risk Analysis.

[32]  Susan A. Murphy,et al.  Monographs on statistics and applied probability , 1990 .

[33]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[34]  Frank E. Harrell,et al.  Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis , 2001 .

[35]  Frank Klawonn,et al.  Learning Graphical Models , 2013 .

[36]  D. Gregori The Susy Safe Project. A web-based registry of foreign bodies injuries in children. , 2006, International journal of pediatric otorhinolaryngology.

[37]  Dario Gregori,et al.  Predicting Severity of Foreign Body Injuries in Children in Upper Airways: An Approach Based on Regression Trees , 2007, Risk analysis : an official publication of the Society for Risk Analysis.

[38]  Oded Maimon,et al.  Evaluation of gene-expression clustering via mutual information distance measure , 2007, BMC Bioinformatics.

[39]  Finn V. Jensen,et al.  Bayesian Networks and Decision Graphs , 2001, Statistics for Engineering and Information Science.

[40]  Eva Pagano,et al.  Using Data Mining Techniques in Monitoring Diabetes Care. The Simpler the Better? , 2011, Journal of Medical Systems.

[41]  P. McCullagh,et al.  Generalized Linear Models , 1984 .