Spatial risk mapping for rare disease with hidden Markov fields and variational EM
Saved in:
| Title: | Spatial risk mapping for rare disease with hidden Markov fields and variational EM |
|---|---|
| Authors: | Forbes, Florence, Garrido, Myriam, Azizi, Lamiae, Doyle, Senan, Abrial, David |
| Contributors: | Azizi, Lamiae, Forbes, Florence, Modelling and Inference of Complex and Structured Stochastic Systems (MISTIS), Centre Inria de l'Université Grenoble Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Centre National de la Recherche Scientifique (CNRS), Unité d'épidémiologie animale, Institut National de la Recherche Agronomique (INRA), INRIA |
| Source: | Ann. Appl. Stat. 7, no. 2 (2013), 1192-1216 |
| Publication Status: | Preprint |
| Publisher Information: | Institute of Mathematical Statistics, 2013. |
| Publication Year: | 2013 |
| Subject Terms: | FOS: Computer and information sciences, ACM: G.: Mathematics of Computing/G.3: PROBABILITY AND STATISTICS, [SDV]Life Sciences [q-bio], disease mapping, BOVINE SPONGIFORM ENCEPHALOPATHY, FRANCE, APPROXIMATIONS, variational EM, Statistics - Applications, 01 natural sciences, BSE, LIKELIHOOD, [STAT.AP] Statistics [stat]/Applications [stat.AP], [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST], FEED, Variational EM, Potts model, Applications (stat.AP), ALGORITHM, 0101 mathematics, [MATH.MATH-ST] Mathematics [math]/Statistics [math.ST], [STAT.AP]Statistics [stat]/Applications [stat.AP], Poisson mixtures, BAN, [STAT.ME] Statistics [stat]/Methodology [stat.ME], VALUES, [STAT.TH] Statistics [stat]/Statistics Theory [stat.TH], Classification, discrete hidden Markov random field, disease mapping, variational EM, BOVINE SPONGIFORM ENCEPHALOPATHY, ALGORITHM, FEED, LIKELIHOOD, FRANCE, BAN, BSE, MIXTURE-MODELS, [STAT.TH]Statistics [stat]/Statistics Theory [stat.TH], Classification, discrete hidden Markov random field, [SDV] Life Sciences [q-bio], Poisson mixtures, Potts model, MIXTURE-MODELS, APPROXIMATIONS, VALUES, Discrete hidden Markov random field, Disease mapping, [STAT.ME]Statistics [stat]/Methodology [stat.ME] |
| Description: | Current risk mapping models for pooled data focus on the estimated risk for each geographical unit. A risk classification, that is, grouping of geographical units with similar risk, is then necessary to easily draw interpretable maps, with clearly delimited zones in which protection measures can be applied. As an illustration, we focus on the Bovine Spongiform Encephalopathy (BSE) disease that threatened the bovine production in Europe and generated drastic cow culling. This example features typical animal disease risk analysis issues with very low risk values, small numbers of observed cases and population sizes that increase the difficulty of an automatic classification. We propose to handle this task in a spatial clustering framework using a nonstandard discrete hidden Markov model prior designed to favor a smooth risk variation. The model parameters are estimated using an EM algorithm and a mean field approximation for which we develop a new initialization strategy appropriate for spatial Poisson mixtures. Using both simulated and our BSE data, we show that our strategy performs well in dealing with low population sizes and accurately determines high risk regions, both in terms of localization and risk level estimation. Published in at http://dx.doi.org/10.1214/13-AOAS629 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org) |
| Document Type: | Article External research report Other literature type Report |
| File Description: | application/pdf |
| ISSN: | 1932-6157 |
| DOI: | 10.1214/13-aoas629 |
| DOI: | 10.48550/arxiv.1312.2800 |
| Access URL: | http://arxiv.org/abs/1312.2800 https://inria.hal.science/inria-00577793v1 https://inria.hal.science/hal-00839184v1/document https://inria.hal.science/hal-00839184v1 https://doi.org/10.1214/13-aoas629 https://hal.inria.fr/inria-00577793 https://hal.archives-ouvertes.fr/hal-00839184 https://hal.inria.fr/hal-00839184/document http://projecteuclid.org/euclid.aoas/1372338484 https://tel.archives-ouvertes.fr/INRIA/hal-00839184v1 https://projecteuclid.org/download/pdfview_1/euclid.aoas/1372338484 https://arxiv.org/pdf/1312.2800.pdf https://arxiv.org/abs/1312.2800 https://ui.adsabs.harvard.edu/abs/2013arXiv1312.2800F/abstract http://projecteuclid.org/euclid.aoas/1372338484 |
| Rights: | implied-oa arXiv Non-Exclusive Distribution |
| Accession Number: | edsair.doi.dedup.....8045c1f2844bde85d6d7d0c22ad4dff6 |
| Database: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstract: | Current risk mapping models for pooled data focus on the estimated risk for each geographical unit. A risk classification, that is, grouping of geographical units with similar risk, is then necessary to easily draw interpretable maps, with clearly delimited zones in which protection measures can be applied. As an illustration, we focus on the Bovine Spongiform Encephalopathy (BSE) disease that threatened the bovine production in Europe and generated drastic cow culling. This example features typical animal disease risk analysis issues with very low risk values, small numbers of observed cases and population sizes that increase the difficulty of an automatic classification. We propose to handle this task in a spatial clustering framework using a nonstandard discrete hidden Markov model prior designed to favor a smooth risk variation. The model parameters are estimated using an EM algorithm and a mean field approximation for which we develop a new initialization strategy appropriate for spatial Poisson mixtures. Using both simulated and our BSE data, we show that our strategy performs well in dealing with low population sizes and accurately determines high risk regions, both in terms of localization and risk level estimation.<br />Published in at http://dx.doi.org/10.1214/13-AOAS629 the Annals of Applied Statistics (http://www.imstat.org/aoas/) by the Institute of Mathematical Statistics (http://www.imstat.org) |
|---|---|
| ISSN: | 19326157 |
| DOI: | 10.1214/13-aoas629 |