Activity and emotion recognition to support early diagnosis of psychiatric diseases

Diagnosis of psychiatric diseases is currently accomplished with questionnaires filled in by the subjects, usually together with a specialist. Such questionnaires are often based on standard, formal scales, where questions range from the ability to cope with household activities, through social interactions, agitation, level of activity, to quality of sleep. In this paper, we propose a context aware framework to support semi-automation of the diagnosis of such diseases, leveraging on authors experience in the fields of activity and emotion recognition. We also introduce a brief description of the main components defining the architecture of the proposed platform and preliminary work on emotion estimation.