Forecasting Sleep Apnea with Dynamic Network Models

Dynamic network models {DNMs} are belief networks for temporal reasoning. The DNM methodology combines techniques from time­ series analysis and probabilistic reasoning to provide (1) a knowledge representation that integrates noncontemporaneous and contem­ poraneous dependencies and (2) methods for iteratively refining these dependencies in re­ sponse to the effects of exogenous influences. We use belief-network inference algorithms to perform forecasting, control, and discrete­ event simulation on DNMs. The belief­ network formulation allows us to move be­ yond the traditional assumptions of linearity in the relationships among time-dependent variables and of normality in their proba­ bility distributions. We demonstrate the DNM methodology on an important forecast­ ing problem in medicine. We conclude with a discussion of how the methodology addresses several limitations found in traditional time­ series analyses.