Dramatic changes in the demographic situation across western countries suggest that new policies towards elderly healthcare should be followed. Among these, timely detection and forecasting of early signs of pathological physical, cognitive or emotional health is of paramount importance, since it will allow for certain medicative strategies to be applied. The USEFIL project as such, aims at developing services that would contribute to the prolongation of seniors' independent living and support them during their daily life activities. An unobtrusive sensor network along with several intelligent processing algorithms will result to a better understanding of the senior population health course. In this paper the modeling and preliminary analysis (first evaluation) of a Decision Support Tool-Subsystem (DSS) for geriatric depression scenario is presented. The decision support tool is based on the construction of a Fuzzy Cognitive Map (FCM) model that describes the geriatric depression scenario. First results are outlined to show the applicability of the proposed methodology. More specifically, a number of virtual senior cases (both healthy and with depressive symptoms) were built with the help of a neuropsychologist. Results of preliminary analysis are promising; however, further expansion of the model with more signs and several other pathologic conditions of seniors such as cognitive impairment and frailty shall be investigated.