Publication: Definición y estudios de redes bayesianas aplicadas a ciencias de la salud y de la vida
Loading...
Official URL
Full text at PDC
Publication Date
2021-09
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
Las redes bayesianas son modelos gráficos probabilísticos que expresan las relaciones de dependencia condicional en un conjunto de variables. Desde su concepción, las redes bayesianas han estado profundamente ligadas a las Ciencias de la Salud y de la Vida, especialmente en el área clínica. Existe una bibliografía extensa sobre aplicaciones de las redes bayesianas a este ámbito. Sin embargo, el análisis de algoritmos de aprendizaje de redes y parámetros, y su aptitud en función de factores como la cantidad de variables, la naturaleza de los datos o la complejidad de la estructura de dependencia no es un tema común en la literatura. En este trabajo, analizamos la aplicación de estas técnicas a problemas descritos en la bibliografía, exploramos el software bnlearn disponible en el lenguaje de programación R documentando nuestro código y evaluamos las estrategias de aprendizaje que mejor se ajustan a cada tipo de datos. Esperamos con ello aportar conocimiento sobre las redes bayesianas y proporcionar un punto de partida para su estudio a profesionales sanitarios e investigadores.
Bayesian networks are graphical probabilistic models which represent conditional dependence relationships among a set of variables. Since their origin, Bayesian networks have been deeply linked to Health and Life Science, especially regarding the clinical field. There is an extensive bibliography about Bayesian networks applications in this field. However, the analysis of algorithms for network learning and their fitness according to factors like the amount of variables, the nature of the data or the complexity of the dependence structure is not a common subject in literature. In this project, we analyze the application of such techniques to problems characterized in the bibliography, we explore the software bnlearn available in the programming language R documenting our code and we evaluate the learning strategies which best fit each kind of data. With this project, we hope to contribute to the knowledge about Bayesian networks and provide a starting point for their study to health professionals and researchers.
Bayesian networks are graphical probabilistic models which represent conditional dependence relationships among a set of variables. Since their origin, Bayesian networks have been deeply linked to Health and Life Science, especially regarding the clinical field. There is an extensive bibliography about Bayesian networks applications in this field. However, the analysis of algorithms for network learning and their fitness according to factors like the amount of variables, the nature of the data or the complexity of the dependence structure is not a common subject in literature. In this project, we analyze the application of such techniques to problems characterized in the bibliography, we explore the software bnlearn available in the programming language R documenting our code and we evaluate the learning strategies which best fit each kind of data. With this project, we hope to contribute to the knowledge about Bayesian networks and provide a starting point for their study to health professionals and researchers.
Description
UCM subjects
Unesco subjects
Keywords
Citation
Albert, J. (2009). Bayesian computation with R. Springer.
Arora, P., Boyne, D., Slater, J. J., Gupta, A., Brenner, D. R., & Druzdzel, M. J. (2019). Bayesian networks for risk prediction using real-world data: a tool for precision medicine. Value in Health, 22(4), 439-445.
Beinlich, I. A., Suermondt, H. J., Chavez, R. M., & Cooper, G. F. (1989). The ALARM monitoring system: A case study with two probabilistic inference techniques for belief networks. In AIME 89 (pp. 247-256). Springer, Berlin, Heidelberg.
Bielza, C., & Larranaga, P. (2014). Bayesian networks in neuroscience: a survey. Frontiers in computational neuroscience, 8, 131.
Córdoba-Sánchez, I., Bielza, C., & Larrañaga, P. (2015). Towards Gaussian Bayesian Network Fusion. In European Conference on Symbolic and Quantitative Approaches to Reasoning and Uncertainty (pp. 519-528). Springer. Correa, E., & Goodacre, R. (2011). A genetic algorithm-Bayesian network approach for the analysis of metabolomics and spectroscopic data: application to the rapid identification of Bacillus spores and classification of Bacillus species. BMC bioinformatics, 12(1), 1-17.
Cover, T. M. (1999). Elements of information theory. John Wiley & Sons.
Dıez, F. J. (1998). Aplicaciones de los modelos gráficos probabilistas en medicina. Sistemas Expertos Probabilísticos, 239-263.
Dimitrova, E. S., Licona, M. P. V., McGee, J., & Laubenbacher, R. (2010). Discretization of time series data. Journal of Computational Biology, 17(6), 853-868
Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2013). Bayesian data analysis. CRC press.
Gorry, G. & G. O. Barnett, G. (1968). Experience with a model of sequential diagnosis. Computers and Biomedical Research, 1, 490-507.
Hansen, K.D., Gentry, J., Long, L., Gentleman, R., Falcon, S., Hahne, F., & Sarkar, D. (2020). Rgraphviz: Provides plotting capabilities for R graph objects. R package version 2.32.0.
Lauritzen S.L., Spiegelhalter D.J. (1988) Local computation with probabilities on graphical structures and their application to expert systems (with discussion). Journal of the Royal Statistical Society. Series B: Statistical Methodology, 50(2), 157–224
Lautner-Csorba, O., Gézsi, A., Semsei, Á. F., Antal, P., Erdélyi, D. J., Schermann, G., ... & Szalai, C. (2012). Candidate gene association study in pediatric acute lymphoblastic leukemia evaluated by Bayesian network based Bayesian multilevel analysis of relevance. BMC medical genomics, 5(1), 1-15.
Lesaffre, E., & Lawson, A. B. (2012). Bayesian biostatistics. John Wiley & Sons.
Mani, S., Valtorta, M., & McDermott, S. (2005). Building Bayesian network models in medicine: The MENTOR experience. Applied Intelligence, 22(2), 93-108.
McCann, R. K., Marcot, B. G., & Ellis, R. (2006). Bayesian belief networks: applications in ecology and natural resource management. Canadian Journal of Forest Research, 36(12), 3053-3062.
Nagarajan, R., Scutari, M., & Lèbre, S. (2013). Bayesian networks in r. Springer, 122, 125-127.
Sachs K, Perez O, Pe’er D, Lauffenburger DA, Nolan GP (2005) Causal protein-signaling networks derived from multiparameter single-cell data. Science 308(5721), 523–529
Scanagatta, M., Salmerón, A., Stella, P. (2019). A survey on Bayesian network structure learning from data. Progress in Artificial Intelligence, 8, 425-489.
Schwartz, S., Baron, J. Y Clarke, J. (1988). A causal Bayesian model for the diagnosis of appendicitis. Uncertainty in Artificial Intelligence, 2, 423-434.
Scutari, M. (2010). Learning Bayesian Networks with the bnlearn R Package. Journal of Statistical Software, 35(3), 1-22. URL http://www.jstatsoft.org/v35/i03/. [Fecha de consulta: 18/05/2021]
Scutari, M., Scutari, M. M., & MMPC, H. P. (2020). Package ‘bnlearn’. Bayesian Network Structure Leaning, parameter learning and inference and inference, R package
Silva, L. C., & Benavides, A. (2001). El enfoque bayesiano: otra manera de inferir. Gaceta Sanitaria, 15(4), 341-346.
Sucar, L. E., & Tonantzintla, M. (2006). Redes bayesianas. Aprendizaje Automático: conceptos básicos y avanzados, 77, 100.
Spiegelhalter, D. & R. P. Knill-Jones, R. (1984). Statistical and knowledge-based approaches to clinical decision support systems, with an application to gastroenterology. Journal of the Royal Statistical Society, 147, 35-77.
Warner, H., Toronto, A., Veasy, L. & Stephenson, R (1961). A mathematical approach to medical diagnosis: Application to congenital heart disease. Journal of the American Medical Association, 177, 177-183.