Multi-level meta-modeling architectures applied to eHealth

Over the last decade, a growing digital universe of unstructured or semi- structured human-sourced information, structured process-mediated data, and well-structured machine-generated data, encourages the adoption of innovative forms of data modeling and information processing to enable enhanced insight, decision making, and process automation applied to a variety of different contexts. Healthcare comprises a notable domain of interest, where the availability of a large amount of information can be exploited to take relevant and tangible benefits in terms of efficiency of the care process, improved out- comes and reduced health system costs. However, due to the complex nature of clinical data, a number of challenges needs to be faced, mainly related on how data characterized by volume, variety, variability, velocity, and veracity can be effectively and efficiently modeled, and how these data can be exploited for increasing the domain knowledge and supporting decision-making processes. The aim of this dissertation is to describe the crucial role played by soft- ware architectures in order to overcome challenges posed by the healthcare context. Specifically, this dissertation addresses the development and applicability of multi-level meta-modeling architectures to various scenarios of eHealth, where flexibility and changeability represent primary requirements. Meta-modeling principles are concretely exploited in the implementation of an adaptable patient-centric Electronic Health Record (EHR) system to face a number of challenging requirements, including: adaptability to different specialities and organizational contexts; run-time configurability by domain experts; interoperability of heterogeneous data produced by various sources and accessed by various actors; applicability of guideline recommendations for evaluating clinical practice compliance; applicability of Activity Recognition techniques for monitoring and classifying human activities in pervasive intelligent environments.