Modular framework to model critical events in stroke patients

Several of the major causes of death in the world are related to neurological diseases, like strokes, sclerosis, or Parkinson's. As a consequence, extraordinary amounts of clinical information are collected. With proper Modeling and Simulation (MS) techniques, predictive models can be defined to help physicians in their diagnoses. We have come to the conclusion that the implementation of an abstract MS methodology to facilitate the processing and modeling of the collected data would be of great help. In this paper, we focus on the development of such an abstract framework, mainly aimed at facilitating and automating the process of data collection and predictive and diagnose models. As a use case, we show how this methodology is applied to determine the stroke type and exitus (i.e. probability of death) of stroke patients in the early stages of their episodes. The best models are evaluated and constantly updated to generate these predictions.