A Bayesian neural network for competing risks models with covariates

C. T C Arsene, P. J G Lisboa, P. Boracchi, E. Biganzoli, M. S H Aung

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a Bayesian neural network for the analysis of competing risk (CR) data model. Based on a previously developed non-linear model namely partial logistic artificial neural network (PLANN) with automatic relevance determination (ARD), this paper proposes an extension for the flexible joint estimation of cause-specific hazards depending on both discrete and continuous covariates (PLANN-CR-ARD) and for censored data. The Bayesian analysis uses Gaussian priors for the neural network parameters and the likelihood function based on the competing risk data is identified as the cross-entropy function. The PLANN-CR-ARD model is illustrated with analyses of an intra-ocular melanoma dataset and comparison with the non-parametric Nelson-Alien estimates of the cause-specific cumulative hazards functions.

Original languageEnglish
Title of host publicationIET Conference Publications
Pages27
Number of pages1
Edition520
DOIs
Publication statusPublished - 2006
EventIET 3rd International Conference MEDSIP 2006: Advances in Medical, Signal and Information Processing - Glasgow, United Kingdom
Duration: Jul 17 2006Jul 19 2006

Other

OtherIET 3rd International Conference MEDSIP 2006: Advances in Medical, Signal and Information Processing
CountryUnited Kingdom
CityGlasgow
Period7/17/067/19/06

Fingerprint

Neural networks
Logistics
Hazards
Data structures
Entropy

Keywords

  • Bayesian neural network
  • Competing risks
  • PLANN-CRARD
  • Survival analysis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Arsene, C. T. C., Lisboa, P. J. G., Boracchi, P., Biganzoli, E., & Aung, M. S. H. (2006). A Bayesian neural network for competing risks models with covariates. In IET Conference Publications (520 ed., pp. 27) https://doi.org/10.1049/cp:20060386

A Bayesian neural network for competing risks models with covariates. / Arsene, C. T C; Lisboa, P. J G; Boracchi, P.; Biganzoli, E.; Aung, M. S H.

IET Conference Publications. 520. ed. 2006. p. 27.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Arsene, CTC, Lisboa, PJG, Boracchi, P, Biganzoli, E & Aung, MSH 2006, A Bayesian neural network for competing risks models with covariates. in IET Conference Publications. 520 edn, pp. 27, IET 3rd International Conference MEDSIP 2006: Advances in Medical, Signal and Information Processing, Glasgow, United Kingdom, 7/17/06. https://doi.org/10.1049/cp:20060386
Arsene CTC, Lisboa PJG, Boracchi P, Biganzoli E, Aung MSH. A Bayesian neural network for competing risks models with covariates. In IET Conference Publications. 520 ed. 2006. p. 27 https://doi.org/10.1049/cp:20060386
Arsene, C. T C ; Lisboa, P. J G ; Boracchi, P. ; Biganzoli, E. ; Aung, M. S H. / A Bayesian neural network for competing risks models with covariates. IET Conference Publications. 520. ed. 2006. pp. 27
@inproceedings{d348fcef70074097bca5522d0f0a1b6b,
title = "A Bayesian neural network for competing risks models with covariates",
abstract = "This paper presents a Bayesian neural network for the analysis of competing risk (CR) data model. Based on a previously developed non-linear model namely partial logistic artificial neural network (PLANN) with automatic relevance determination (ARD), this paper proposes an extension for the flexible joint estimation of cause-specific hazards depending on both discrete and continuous covariates (PLANN-CR-ARD) and for censored data. The Bayesian analysis uses Gaussian priors for the neural network parameters and the likelihood function based on the competing risk data is identified as the cross-entropy function. The PLANN-CR-ARD model is illustrated with analyses of an intra-ocular melanoma dataset and comparison with the non-parametric Nelson-Alien estimates of the cause-specific cumulative hazards functions.",
keywords = "Bayesian neural network, Competing risks, PLANN-CRARD, Survival analysis",
author = "Arsene, {C. T C} and Lisboa, {P. J G} and P. Boracchi and E. Biganzoli and Aung, {M. S H}",
year = "2006",
doi = "10.1049/cp:20060386",
language = "English",
isbn = "0863416586",
pages = "27",
booktitle = "IET Conference Publications",
edition = "520",

}

TY - GEN

T1 - A Bayesian neural network for competing risks models with covariates

AU - Arsene, C. T C

AU - Lisboa, P. J G

AU - Boracchi, P.

AU - Biganzoli, E.

AU - Aung, M. S H

PY - 2006

Y1 - 2006

N2 - This paper presents a Bayesian neural network for the analysis of competing risk (CR) data model. Based on a previously developed non-linear model namely partial logistic artificial neural network (PLANN) with automatic relevance determination (ARD), this paper proposes an extension for the flexible joint estimation of cause-specific hazards depending on both discrete and continuous covariates (PLANN-CR-ARD) and for censored data. The Bayesian analysis uses Gaussian priors for the neural network parameters and the likelihood function based on the competing risk data is identified as the cross-entropy function. The PLANN-CR-ARD model is illustrated with analyses of an intra-ocular melanoma dataset and comparison with the non-parametric Nelson-Alien estimates of the cause-specific cumulative hazards functions.

AB - This paper presents a Bayesian neural network for the analysis of competing risk (CR) data model. Based on a previously developed non-linear model namely partial logistic artificial neural network (PLANN) with automatic relevance determination (ARD), this paper proposes an extension for the flexible joint estimation of cause-specific hazards depending on both discrete and continuous covariates (PLANN-CR-ARD) and for censored data. The Bayesian analysis uses Gaussian priors for the neural network parameters and the likelihood function based on the competing risk data is identified as the cross-entropy function. The PLANN-CR-ARD model is illustrated with analyses of an intra-ocular melanoma dataset and comparison with the non-parametric Nelson-Alien estimates of the cause-specific cumulative hazards functions.

KW - Bayesian neural network

KW - Competing risks

KW - PLANN-CRARD

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=34047142243&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34047142243&partnerID=8YFLogxK

U2 - 10.1049/cp:20060386

DO - 10.1049/cp:20060386

M3 - Conference contribution

AN - SCOPUS:34047142243

SN - 0863416586

SN - 9780863416583

SP - 27

BT - IET Conference Publications

ER -