Inpatient admissions in different wards/clinics between 1998 and 2014 are considered. A Bayesian network (BN) structure is estimated, directly, from data in order to compute the joint probabilities of the different patient profiles as one of the main objective is to identify the most probable configurations of the wards/clinics. Knowing which wards/clinics are more interrelated could be useful for a better organization of the hospital. Once the Bayesian network is estimated, evidence for some nodes (in our case the history of a patient up to a certain stage) can be propagated through the graph, and the BN shows how such evidence changes the marginal distributions of the remaining nodes. Therefore, it is possible to predict in which ward/clinic there will be a next admission, assuming that there is one. Cross-validation are also performed to test the predictive ability of the BN.

Conigliani, C., Petitti, T., & Vitale, V. (2016). Bayesian networks for the analysis of inpatient admissions. In 9th International Conference on Computational and Methodological Statistics.

Bayesian networks for the analysis of inpatient admissions

CONIGLIANI, CATERINA;VITALE, VINCENZINA
2016

Abstract

Inpatient admissions in different wards/clinics between 1998 and 2014 are considered. A Bayesian network (BN) structure is estimated, directly, from data in order to compute the joint probabilities of the different patient profiles as one of the main objective is to identify the most probable configurations of the wards/clinics. Knowing which wards/clinics are more interrelated could be useful for a better organization of the hospital. Once the Bayesian network is estimated, evidence for some nodes (in our case the history of a patient up to a certain stage) can be propagated through the graph, and the BN shows how such evidence changes the marginal distributions of the remaining nodes. Therefore, it is possible to predict in which ward/clinic there will be a next admission, assuming that there is one. Cross-validation are also performed to test the predictive ability of the BN.
978-9963-2227-1-1
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11590/311552
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact