|
Activity Number:
|
383
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Risk Analysis
|
| Abstract - #310262 |
|
Title:
|
A Bayesian Network-Driven Approach for Quantitative Risk Assessment of Foreign Body Injuries in Children
|
|
Author(s):
|
Cecilia Scarinzi*+ and Paola Berchialla and Silvia Snidero and Alexander Stancu and Roberto Corradetti and Dario Gregori
|
|
Companies:
|
Università degli Studi di Torino and University of Torino and S&A S.r.l. and University of Torino and University of Torino and University of Torino
|
|
Address:
|
piazza arbarello 8, Torino, 10121, Italy
|
|
Keywords:
|
Bayesian network ; children ; foreign body injuries ; quantitative risk assessment
|
|
Abstract:
|
Injuries due to foreign body (FB) aspiration/ingestion among children represent a common public heath issue. The aim of this study is to provide a Bayesian Network (BN) that allows for the identification and quantification of risk factors for FB injuries. Combining in a coherent way qualitative and quantitative data, a BN learning algorithm was used to generate a network of the complex relationships among causal factors, both in its topology and its distribution, followed by model verification. Thus, the BN was used for drawing inference on propositions of interest like the probability of injured children to experience complications or hospitalization. Using BN is a new method to evaluate unknown outcomes. This tool of prediction analysis may assist in determining potential hazard of object characteristics. Previous statistical analyses do not allow individual risk analysis.
|
