Facial expressions are visible signs of the affective and psychological state of a person, which is strictly correlated with the pathogenesis of clinically relevant diseases and more in general with individuals' wellbeing. The main idea highlighted in this paper is the exploitation of the facial expression analysis for wellbeing monitoring and self-management. This will occur by an innovative multisensory device that will be able to collect images and signals, extract quantitative features of facial expression related to stress, anxiety and fatigue and map them to computational descriptors of an individual's wellbeing. The latter phase will be based on a virtual individual's model conceived to allow the computation and tracing of the daily evolution of individual's wellness. Personalized advices and coaching messages will support the user in keeping a healthy lifestyle and counteract potentially harmful behaviours. The work is part of the FP7 STREP SEMEOTICONS project whose application field will be the prevention of cardio-metabolic risk, for which healthcare systems are registering an exponential growth of social costs.