Bayes Net Modeling
Joseph A. Tatman, Barry Charles Ezell · 2016
Bayesian networks (BNs) have many unique advantages for diagnostic and other problems in medicine. This chapter covers four application areas of Bayesian networks include automation of patient aids, clinical diagnosis and treatment, medical image interpretation, and public health. Dynamic Bayesian Networks (DBN) support modeling changes in patients’ condition over time due to both diseases and treatments, using probabilistic relationships between different clinical variables, both within and across different points in time. Researchers used BNs to discover relations between genes, environment, and disease. They applied their approach to a study of bladder cancer in the United States. An advantage of BNs over the more standard logistic regression approaches includes the ability to infer unknown values of some nodes given other nodes with known values. BNks facilitate causal models that a human user can more easily understand, interact with and explore.