Abstract
Learning from electronic medical records (EMR) poses many challenges from a knowledge representation point of view. This chapter focuses on how to cope with two specific challenges: the relational nature of EMRs and the uncertain dependence between a patient’s past and future health status. We discuss three different approaches for allowing standard propositional learners to incorporate relational information. We evaluate these approaches on three real-world tasks where the goal is to use EMRs to predict whether a patient will have an adverse reaction to a medication.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A Datalog clause is non-recursive by definition if the predicate appearing in its head does not appear in its body. A Datalog program, or theory, is non-recursive if all its clauses are non-recursive.
- 2.
In principle, VISTA can use any evaluation metric to evaluate the quality of the model such as (conditional) likelihood, accuracy, or ROC analysis.
References
Alphonse, E., Rouveirol, C.: Lazy propositionalisation for relational learning. In: 14th European Conference on Artificial Intelligence, pp. 256–260. IOS Press (2000)
Boyd, K., Davis, J., Page, D., Costa, V.S.: Unachievable region in precision-recall space and its effect on empirical evaluation. In: Proceedings of the 29th International Conference on Machine Learning. Omnipress (2012)
Chaudhuri, S., Dayal, U.: An overview of data warehousing and OLAP technology. SIGMOD Rec. 26(1), 65–74 (1997)
Davis, J., Ong, I., Struyf, J., Burnside, E., Page, D., Costa, V.S.: Change of representation for statistical relational learning. In: Proceedings of the 20th International Joint Conference on Artificial Intelligence, pp. 2719–2726 (2007)
Davis, J., Goadrich, M.: The relationship between precision-recall and ROC curves. In: Proceedings of the 23rd International Conference on Machine learning, pp. 233–240. ACM Press (2006)
Friedman, N., Geiger, D., Goldszmidt, M.: Bayesian networks classifiers. Mach. Learn. 29, 131–163 (1997)
Kearney, P., Baigent, C., Godwin, J., Halls, H., Emberson, J., Patrono, C.: Do selective cyclo-oxygenase-2 inhibitors and traditional non-steroidal anti-inflammatory drugs increase the risk of atherothrombosis? meta-analysis of randomised trials. BMJ 332, 1302–1308 (2006)
Kramer, S., Lavrac, N., Flach, P.: Propositionalization approaches to relational data mining. In: Džeroski, S., Lavrac, N. (eds.) Relational Data Mining Part III, pp. 262–291. Springer, Heidelberg (2001)
Landwehr, N., Kersting, K., and Raedt, L. D. nFOIL: Integrating Naive Bayes and FOIL. In: Proceeding of the 20th National Conference on Artificial Intelligence, pp. 795–800 (2005)
Landwehr, N., Passerini, A., Raedt, L.D., Frasconi, P.: kFOIL: Learning simple relational kernels. In: Proceedings of the 21st National Conference on Artificial Intelligence (2006)
Lavrač, N., Džeroski, S. (eds.): Relational Data Mining. Springer, Heidelberg (2001)
Lavrač, N., Džeroski, S.: Inductive learning of relations from noisy examples. In: Muggleton, S. (ed.) Inductive Logic Programming, pp. 495–516. Academic Press, London (1992)
McCarty, C., Wilke, R., Giampietro, P., Wesbrook, S., Caldwell, M.: Personalized Medicine Research Project (PMRP): design, methods and recruitment for a large population-based biobank. Personalized Med. 2, 49–79 (2005)
Muggleton, S.: Inverse entailment and Progol. New Gener. Comput. 13, 245–286 (1995)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Mateo, California (1988)
Platt, R., Carnahan, R.: The US food and drug administration’s mini-sentinel program. Pharmacoepidemiol. Drug Saf. 21, 1–303 (2012)
Pompe, U., Kononenko, I.: Naive Bayesian classifier within ILP-R. In: De Raedt, L. (ed.), Proceedings of the 5th International Conference on Inductive Logic Programming, pp. 417–436 (1995)
Popescul, A., Ungar, L., Lawrence, S., Pennock, D. Statistical relational learning for document mining. In: Proceeding of the 3rd International Conference on Data Mining, pp. 275–282 (2003)
Quinlan, J.R.: Induction of decision trees. Mach. Learn. 1, 81–106 (1986)
Quinlan, J.R.: Learning logical definitions from relations. Mach. Learn. 5, 239–266 (1990)
Srinivasan, A.: The Aleph Manual (2001)
Vapnik, V.: The Nature of Statistical Learning Theory. Information Science and Statistics. Springer, Heidelberg (1999)
Acknowledgements
JD is partially supported by the Research Fund K.U.Leuven (CREA/11/015 and OT/11/051), EU FP7 Marie Curie Career Integration Grant (\(\#\)294068) and FWO-Vlaanderen (G.0356.12). VSC is funded by ERDF through Programme COMPETE and by the Portuguese Government through FCT Foundation for Science and Technology projects LEAP (PTDC/EIA-CCO/112158/2009) and ADE (PTDC/EIA-EIA/121686/2010). MC, PP, EB and DP gratefully acknowledge the support of NIGMS grant R01GM097618-01.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Davis, J., Costa, V.S., Peissig, P., Caldwell, M., Page, D. (2015). Predicting Adverse Drug Events from Electronic Medical Records. In: Hommersom, A., Lucas, P. (eds) Foundations of Biomedical Knowledge Representation. Lecture Notes in Computer Science(), vol 9521. Springer, Cham. https://doi.org/10.1007/978-3-319-28007-3_16
Download citation
DOI: https://doi.org/10.1007/978-3-319-28007-3_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-28006-6
Online ISBN: 978-3-319-28007-3
eBook Packages: Computer ScienceComputer Science (R0)