A reference architecture for personalized and self-adaptive e-health apps

A wealth of e-Health mobile apps are available for many purposes, such as life style improvement, mental coaching, etc. The interventions, prompts, and encouragements of e-Health apps sometimes take context into account (e.g., previous interactions or geographical location of the user), but they still tend to be rigid, e.g., by using fixed rule sets or being not sufficiently tailored towards individuals. Personalization to the different users’ characteristics and run-time adaptation to their changing needs and context provide a great opportunity for getting users continuously engaged and active, eventually leading to better physical and mental conditions. This paper presents a reference architecture for enabling AI-based personalization and self-adaptation of mobile apps for e-Health. The reference architecture makes use of multiple MAPE loops operating at different levels of granularity and for different purposes.