Learning T2D Evolving Complexity from EMR and Administrative Data by Means of Continuous Time Bayesian Networks

Simone Marini, Arianna Dagliati, Lucia Sacchi, Riccardo Bellazzi

2016

Abstract

Predicting the complexity level (i.e. the number of complications and their related hospitalizations) in a T2D cohort is a critical step in prevention, resource optimization and overall patient management. Our data set was obtained by monitoring a T2D diabetic cohort along up to 10 years through electronic medical records of a local healthcare agency data warehouse. In order to conveniently handle temporarily sparse data, we designed a model describing the cohort evolution with Continuous Time Bayesian Networks (CTBN). The network structure and its parameters are entirely data driven. Compared to traditional Bayesian Networks, CTBNs admit cycles. As consequence, CTBNs fit the complexity of chronic metabolic syndromes where variables show a reciprocal influence. Network nodes represent metabolic (glycated hemoglobin, lipid profile (cholesterol, triglycerides), and biometric (BMI) data. We observed how these variables directly or indirectly affect the disease level of complexity, and how the variables influence the cumulative adverse events a patient undergoes.

References

  1. Acerbi, E., Stella, F. 2014. Continuous Time Bayesian Networks for Gene Network Reconstruction: A Comparative Study on Time Course Data. Bioinformatics Research and Applications. Springer International Publishing, 176-187.
  2. American Diabetes Association 2013. Standards of medical care in diabetes. Diabetes Care 36 (1): S11- S66.
  3. Dagliati, A., Sacchi, L., Bucalo, M., Segagni, D., Zarkogianni, K., Martinez Millana, A., Cancela, J., Sambo, F., Fico, G., Meneu Barreira, M.T., Cerra, C., Nikita, K., Cobelli, C., Chiovato, L., Arredondo, M.T., Bellazzi, R. 2014a. A Data Gathering Framework to Collect Type 2 Diabetes Patients Data. Biomedical and Health Informatics (BHI), 2014 IEEE-EMBS International Conference Proceedings. 244 - 247.
  4. Dagliati A., Sacchi, L., Cerra, C., Leporati, P., De Cata, P., Chiovato, L., Holmes, J.H., Bellazzi, R. 2014b. Temporal Data Mining and Process Mining Techniques to Identify Cardiovascular RiskAssociated Clinical Pathways in Type 2 Diabetes Patients Biomedical and Health Informatics (BHI), 2014 IEEE-EMBS International Conference Proceedings. 240 - 243.
  5. Gatti, E., Luciani, D., Stella, F. 2012. A continuous time Bayesian network model for cardiogenic heart failure. Flexible Services and Manufacturing Journal 4(4), 496-515.
  6. International Diabetes Federation. 2014. IDF Diabetes Atlas. 6th edn, 2014 Update. Brussels, Belgium: International Diabetes Federation.
  7. Liu, B., Thiagarajan, P.S., Hsu, D. 2009. Probabilistic Approximations of Signaling Pathway Dynamics. Computational Methods in Systems Biology. Springer Berlin Heidelberg.
  8. McEwan, P., Foos, V., Palmer, J.L., Lamotte, M., Lloyd, A., Grant, D. 20014. Validation of the IMS CORE Diabetes Model. Value Health (6):714-24.
  9. Marini, S., Trifoglio, E., Barbarini, N., Sambo, F., Di Camillo, B., Malovini, A., Manfrini, M., Cobelli, C., Bellazzi, R. 2015. A Dynamic Bayesian Network model for long-term simulation of clinical complications in type 1 diabetes. Journal of Biomedical Informatics, ePub ahead of print.
  10. Nodelman, U., Shelton, C.R., Koller, D. 2002a. Continuous time bayesian networks. UAI02 Proceedings, 378-387.
  11. Nodelman, U., Shelton, C.R., and Koller, D. 2002b. Learning continuous time bayesian networks. In Proc. of the 19th Conf. on Uncertainty in Artificial Intelligence, pages 451-458.
  12. Shelton, C.R., Fan, Y., Lam, W., Lee, J., Xu. J. 2010. Continuous Time Bayesian Network Reasoning and Learning Engine. The Journal of Machine Learning Research 11: 1137-1140.
  13. Solano, M.P., Goldberg, R.B. 2006. Lipid management in type 2 diabetes. Clinical Diabetes 24(1): 27-32.
  14. Tarride, J.E., Hopkins, R., Blackhouse, G., Bowen, J.M., Bischof, M., Von Key-serlingk, C., O'Reilly, D., Xie, F., Goeree, R. 2010. A review of methods used in longterm cost-effectiveness models of diabetes mellitus treatment. Pharmacoeconomics. 28(4):255-77.
  15. Wang, X., Sontag, D., Wang, F. 2014. Unsupervised Learning of Disease Progression Models. Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining, 85-94.
Download


Paper Citation


in Harvard Style

Marini S., Dagliati A., Sacchi L. and Bellazzi R. (2016). Learning T2D Evolving Complexity from EMR and Administrative Data by Means of Continuous Time Bayesian Networks . In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF, (BIOSTEC 2016) ISBN 978-989-758-170-0, pages 338-344. DOI: 10.5220/0005708103380344


in Bibtex Style

@conference{healthinf16,
author={Simone Marini and Arianna Dagliati and Lucia Sacchi and Riccardo Bellazzi},
title={Learning T2D Evolving Complexity from EMR and Administrative Data by Means of Continuous Time Bayesian Networks},
booktitle={Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF, (BIOSTEC 2016)},
year={2016},
pages={338-344},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005708103380344},
isbn={978-989-758-170-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF, (BIOSTEC 2016)
TI - Learning T2D Evolving Complexity from EMR and Administrative Data by Means of Continuous Time Bayesian Networks
SN - 978-989-758-170-0
AU - Marini S.
AU - Dagliati A.
AU - Sacchi L.
AU - Bellazzi R.
PY - 2016
SP - 338
EP - 344
DO - 10.5220/0005708103380344