Joint Bayesian Modelling of Internal Dependencies and Relevant Multimorbidities of a Heterogeneous Disease

Péter Marx, András Millinghoffer, Gabriella Juhász, Péter Antal
Proceedings of the Eighth International Conference on Probabilistic Graphical Models, PMLR 52:310-320, 2016.

Abstract

A heterogeneous target disease represented by multiple descriptors and disease subtypes frequently has a rich internal dependency structure. The identification of comorbidities and particularly the multimorbidities of such diseases requires very large sample size as relevant comorbidities may form complex interactions. We demonstrate this phenomena by applying a Bayesian probabilistic graphical model on a large-scale medical datasets UK Biobank (117,392 samples), specifically by showing that in this case the posterior landscape of multimorbidities is still flat. As a potential solution, we evaluate a Bayesian method, which provides a hierarchic, multivariate characterization of strongly relevant morbidities and a Bayesian, systems-based score for exploring interactions for a heterogeneous disease. It explores complete sets of strongly relevant comorbidities using full multivariate representation for the internal dependencies within the target disease. We used depression as target, a heterogeneous disease in the UK Biobank dataset. Results are compared against scenarios using a univariate and an independent, multivariate representation of the target medical condition, specifically investigating multitarget interaction posteriors and its approximations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v52-marx16, title = {Joint {B}ayesian Modelling of Internal Dependencies and Relevant Multimorbidities of a Heterogeneous Disease}, author = {Marx, Péter and Millinghoffer, András and Juhász, Gabriella and Antal, Péter}, booktitle = {Proceedings of the Eighth International Conference on Probabilistic Graphical Models}, pages = {310--320}, year = {2016}, editor = {Antonucci, Alessandro and Corani, Giorgio and Campos}, Cassio Polpo}, volume = {52}, series = {Proceedings of Machine Learning Research}, address = {Lugano, Switzerland}, month = {06--09 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v52/marx16.pdf}, url = {https://proceedings.mlr.press/v52/marx16.html}, abstract = {A heterogeneous target disease represented by multiple descriptors and disease subtypes frequently has a rich internal dependency structure. The identification of comorbidities and particularly the multimorbidities of such diseases requires very large sample size as relevant comorbidities may form complex interactions. We demonstrate this phenomena by applying a Bayesian probabilistic graphical model on a large-scale medical datasets UK Biobank (117,392 samples), specifically by showing that in this case the posterior landscape of multimorbidities is still flat. As a potential solution, we evaluate a Bayesian method, which provides a hierarchic, multivariate characterization of strongly relevant morbidities and a Bayesian, systems-based score for exploring interactions for a heterogeneous disease. It explores complete sets of strongly relevant comorbidities using full multivariate representation for the internal dependencies within the target disease. We used depression as target, a heterogeneous disease in the UK Biobank dataset. Results are compared against scenarios using a univariate and an independent, multivariate representation of the target medical condition, specifically investigating multitarget interaction posteriors and its approximations.} }
Endnote
%0 Conference Paper %T Joint Bayesian Modelling of Internal Dependencies and Relevant Multimorbidities of a Heterogeneous Disease %A Péter Marx %A András Millinghoffer %A Gabriella Juhász %A Péter Antal %B Proceedings of the Eighth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2016 %E Alessandro Antonucci %E Giorgio Corani %E Cassio Polpo Campos} %F pmlr-v52-marx16 %I PMLR %P 310--320 %U https://proceedings.mlr.press/v52/marx16.html %V 52 %X A heterogeneous target disease represented by multiple descriptors and disease subtypes frequently has a rich internal dependency structure. The identification of comorbidities and particularly the multimorbidities of such diseases requires very large sample size as relevant comorbidities may form complex interactions. We demonstrate this phenomena by applying a Bayesian probabilistic graphical model on a large-scale medical datasets UK Biobank (117,392 samples), specifically by showing that in this case the posterior landscape of multimorbidities is still flat. As a potential solution, we evaluate a Bayesian method, which provides a hierarchic, multivariate characterization of strongly relevant morbidities and a Bayesian, systems-based score for exploring interactions for a heterogeneous disease. It explores complete sets of strongly relevant comorbidities using full multivariate representation for the internal dependencies within the target disease. We used depression as target, a heterogeneous disease in the UK Biobank dataset. Results are compared against scenarios using a univariate and an independent, multivariate representation of the target medical condition, specifically investigating multitarget interaction posteriors and its approximations.
RIS
TY - CPAPER TI - Joint Bayesian Modelling of Internal Dependencies and Relevant Multimorbidities of a Heterogeneous Disease AU - Péter Marx AU - András Millinghoffer AU - Gabriella Juhász AU - Péter Antal BT - Proceedings of the Eighth International Conference on Probabilistic Graphical Models DA - 2016/08/15 ED - Alessandro Antonucci ED - Giorgio Corani ED - Cassio Polpo Campos} ID - pmlr-v52-marx16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 52 SP - 310 EP - 320 L1 - http://proceedings.mlr.press/v52/marx16.pdf UR - https://proceedings.mlr.press/v52/marx16.html AB - A heterogeneous target disease represented by multiple descriptors and disease subtypes frequently has a rich internal dependency structure. The identification of comorbidities and particularly the multimorbidities of such diseases requires very large sample size as relevant comorbidities may form complex interactions. We demonstrate this phenomena by applying a Bayesian probabilistic graphical model on a large-scale medical datasets UK Biobank (117,392 samples), specifically by showing that in this case the posterior landscape of multimorbidities is still flat. As a potential solution, we evaluate a Bayesian method, which provides a hierarchic, multivariate characterization of strongly relevant morbidities and a Bayesian, systems-based score for exploring interactions for a heterogeneous disease. It explores complete sets of strongly relevant comorbidities using full multivariate representation for the internal dependencies within the target disease. We used depression as target, a heterogeneous disease in the UK Biobank dataset. Results are compared against scenarios using a univariate and an independent, multivariate representation of the target medical condition, specifically investigating multitarget interaction posteriors and its approximations. ER -
APA
Marx, P., Millinghoffer, A., Juhász, G. & Antal, P.. (2016). Joint Bayesian Modelling of Internal Dependencies and Relevant Multimorbidities of a Heterogeneous Disease. Proceedings of the Eighth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 52:310-320 Available from https://proceedings.mlr.press/v52/marx16.html.

Related Material