Bayesian network classifier for medical data analysis

Int. J. of Computers, Communications and …, 2008

Bayesian Networks encode causal relations between variables using probability and graph theory. We exploit the causal relations to detect dependency structures in database consisting of a small number of observations having high dimensions. Bypass surgery survival chance must be inferred from a database consisting of 66 medical examinations for 313 patients. Tree-like Bayesian were inferred based on mutual information and than analysed for classification of data, respect to survival. Bayesian Network approach allows us to interpret the predictions of the system thus helping the doctor in after surgery treatment prescription. In this paper we present the used methods and results on artificial data.

2011

With the current trend toward pervasive health care, personalised health care, and the ever growing amount of evidence coming from biomedical research, methods that can handle reasoning and learning under uncertainty are becoming more and more important. The ongoing developments of the past two decades in the field of artificial intelligence have made it now possible to apply probabilistic methods to solve problems in real-world biomedical domains. Many representations have been suggested for solving problems in biomedical domains. Bayesian networks and influence diagrams have proved themselves useful for problems where probabilistic uncertainty is important, such as medical decision making and prognostics; logics have proved themselves useful in areas such as diagnosis. In recent years, the field of statistical relational learning has led to new formalisms which integrate probabilistic graphical models and logic. These formalisms provide exciting new opportunities for medical applications as they can be used to learn from structured medical data and reason with them using both logical and probabilistic methods. Another major theme for this workshop is in the handling of semantic concepts such as space and time in the biomedical domain. Space is an important concept when developing probabilistic models of, e.g., the spread of infectious disease, either in the hospital or in the community at large. Temporal reasoning is especially important in the context of personalised health care. Consider for example the translation of biomedical research that is expected to lead to more complex decision making, e.g., how to optimally select a sequence of drugs targeting biological pathways when treating a malignant tumour. There are strong expectations that such personalised and specific drugs will soon be available in the clinical practice. We selected eleven papers for full presentation. All these contributions fit the format of the workshop: they develop new approaches for integrating logical and semantical concepts with probabilistic methods or apply existing methods to problems from the biomedical domain. Furthermore, we feel honoured to have Jesse Davis and Milos Hauskrecht as invited speakers. Jesse Davis has made significant contributions in the application of statistical relational learning techniques in the medical domain. Milos Hauskrecht is well-known for his work in the analysis of time-series data (e.g., using POMDPs) in biomedical informatics. The organisers would like to acknowledge the support from the AIME organisation. We would also like to thank the program committee members for their support and reviewing, which have improved the accepted papers significantly.

Cancer Informatics, 2014

The purpose of this investigation is to develop and evaluate a new Bayesian network (BN)-based patient survivorship prediction method. The central hypothesis is that the method predicts patient survivorship well, while having the capability to handle high-dimensional data and be incorporated into a clinical decision support system (CDSS). We have developed EBMC_Survivorship (EBMC_S), which predicts survivorship for each year individually. EBMC_S is based on the EBMC BN algorithm, which has been shown to handle high-dimensional data. BNs have excellent architecture for decision support systems. In this study, we evaluate EBMC_S using the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) dataset, which concerns breast tumors. A 5-fold cross-validation study indicates that EMBC_S performs better than the Cox proportional hazard model and is comparable to the random survival forest method. We show that EBMC_S provides additional information such as sensitivity analyses, which covariates predict each year, and yearly areas under the ROC curve (AUROCs). We conclude that our investigation supports the central hypothesis.

2009

Bayesian Networks represent one of the most successful tools for medical diagnosis and therapies follow-up. We present an algorithm for Bayesian network structure learning, that is a variation of the standard search-and-score approach. The proposed approach overcomes the creation of redundant network structures that may include non significant connections between variables. In particular, the algorithm finds which relationships between the variables must be prevented, by exploiting the binarization of a square matrix containing the mutual information (MI) among all pairs of variables. Four different binarization methods are implemented. The MI binary matrix is exploited as a preconditioning step for the subsequent greedy search procedure that optimizes the network score, reducing the number of possible search paths in the greedy search. Our approach has been tested on two different medical datasets and compared against the standard search-and-score algorithm as implemented in the DEAL package.

2012

This thesis investigates the use of Bayesian Networks (BNs), augmented by the Dynamic Discretization Algorithm, to model a variety of clinical problems. In particular, the thesis demonstrates four novel applications of BN and dynamic discretization to clinical problems. Firstly, it demonstrates the flexibility of the Dynamic Discretization Algorithm in modeling existing medical knowledge using appropriate statistical distributions. Many practical applications of BNs use the relative frequency approach while translating existing medical knowledge to a prior distribution in a BN model. This approach does not capture the full uncertainty surrounding the prior knowledge. Secondly, it demonstrates a novel use of the multinomial BN formulation in learning parameters of categorical variables. The traditional approach requires fixed number of parameters during the learning process but this framework allows an analyst to generate a multinomial BN model based on the number of parameters requi...

International Journal of Modern Physics C, 2006

A Bayesian network is a directed acyclic graph in which each node represents a variable and each arc a probabilistic dependency; they are used to provide: a compact form to represent the knowledge and flexible methods of reasoning. Obtaining it from data is a learning process that is divided in two steps: structural learning and parametric learning. In this paper we define an automatic learning method that optimizes the Bayesian networks applied to classification, using a hybrid method of learning that combines the advantages of the induction techniques of the decision trees (TDIDT-C4.5) with those of the Bayesian networks. The resulting method is applied to prediction in health domain.

Lecture Notes in Computer Science, 2009

While for many problems in medicine classification models are being developed, Bayesian network classifiers do not seem to have become as widely accepted within the medical community as logistic regression models. We compare first-order logistic regression and naive Bayesian classification in the domain of reproductive medicine and demonstrate that the two techniques can result in models of comparable performance. For Bayesian network classifiers to become more widely accepted within the medical community then, we feel that they should be better aligned with their context of application. We describe how to incorporate well-known concepts of clinical relevance in the process of constructing and evaluating Bayesian network classifiers to achieve such an alignment.

Diagnostic Cytopathology, 2018

BackgroundIn the era of extensive data collection, there is a growing need for a large scale data analysis with tools that can handle many variables in one modeling framework. In this article, we present our recent applications of Bayesian network modeling to pathology informatics.MethodsBayesian networks (BNs) are probabilistic graphical models that represent domain knowledge and allow investigators to process this knowledge following sound rules of probability theory. BNs can be built based on expert opinion as well as learned from accumulating data sets. BN modeling is now recognized as a suitable approach for knowledge representation and reasoning under uncertainty. Over the last two decades BN have been successfully applied to many studies on medical prognosis and diagnosis.ResultsBased on data and expert knowledge, we have constructed several BN models to assess patient risk for subsequent specific histopathologic diagnoses and their related prognosis in gynecological cytopath...

Artificial Intelligence in Medicine, 2004

Due to the uncertainty of many of the factors that influence the performance of an emergency medical service, we propose using Bayesian networks to model this kind of system. We use different algorithms for learning Bayesian networks in order to build several models, from the hospital manager's point of view, and apply them to the specific case of the emergency service of a Spanish hospital. This first study of a real problem includes preliminary data processing, the experiments carried out, the comparison of the algorithms from different perspectives, and some potential uses of Bayesian networks for management problems in the health service.

Applied Artificial Intelligence, 1989

This paper relates our experience in developing a mechanism for reasoning about the di erential diagnosis of cases involving the symptoms of heart failure using a causal model of the cardiovascular hemodynamics with probabilities relating cause to e ect. Since the problem requires the determination of causal mechanism as well as primary cause, the model has many intermediate nodes as well as causal circularities requiring a heuristic approach to evaluating probabilities. The method we have developed builds hypotheses incrementally by adding the highest probability path to each nding to the hypothesis. With a number of enhancements and computational tactics, this method has proven e ective for generating good hypotheses for typical cases in less than a minute.