Methods for acquisition and integration of personal wellness parameters

Wellness indicates the state or condition of being in good physical and mental health. Stress is a common state of emotional strain that plays a crucial role in the everyday quality of life. Nowadays, there is a growing individual awareness of the importance of a proper lifestyle and a generalized trend to become an active part in monitoring, preserving, and improving personal wellness for both physical and emotional aspects. The majority studies in this field relies on the evaluation of the changes of sensed parameters passing from rest to “maximal” stress. However, the vast majority of people usually experiences stressing circumstances in everyday life. This led us to investigate the impact of mild cognitive activation which can be somehow comparable to usual situations that everyone can face in daily life. Several signals and data can be useful to characterize the state of a person, but not all of them are equally important. So it is crucial to analyse the mutual relevance of the different pieces of information. In this work we focus on a subset of well-established psychophysical descriptors and we identified a set of devices enabling the measurement of these parameters . The design of the experimental setup and the selection of sensing devices were driven by qualitative criteria such as intrusiveness, reliability, and ease of use. These are deemed crucial for implementing effective (self-)monitoring strategies. A reference dataset, named “Mild Cognitive Activation” (MCA), was collected. The last aim of the project was the definition of a quantitative model for data integration providing a concise description of the wellness status of a person. This process was based on unsupervised learning paradigms. Data from MCA were integrated with data from the “Stress Recognition in Automobile Drivers” dataset . This allowed a cross validation of the integration methodology.