This work aims to assess the predictive value of physiological data, daily collected in a telemonitoring study, in the early detection of heart failure decompensation events. The main hypothesis is that physiological time series with similar progression (trends) may have prognostic value in future clinical states (decompensation or normal conditions).
The strategy is composed of two main steps: a trend similarity analysis followed by a predictive procedure. Basically, founded on the trend similarity measure, a set of time series presenting a progression similar with the current condition is identified in the historical data set, which is then employed, through a nearest neighbour approach, in the current prediction (decompensation event or normal condition). The proposed strategy is validated using physiological data collected during the myHeart telemonitoring study. The obtained results suggest, in general, that the physiological data have predictive value and, in particular, that the proposed similarity scheme is particularly appropriate to address the early detection of heart failure decompensation.