An iintegrated algorithmic and software solution for biological rhythms analysis: Application to long-term data series

Here we introduce an algorithmic tool to extract significant information out of the long-term physiological monitoring data following an originally designed methodology that is implemented as an integrative software solution. The algorithm consists of several consecutive steps including elimination of outlier measurements, filtering out high-frequency noise, detection of the long-term trend and an enhanced elicitation and analysis procedure of the rhythms based on the spectral analysis of the fluctuations around the detected trend. The designed solutions are especially suited to operate with non-equidistant data series that are common in ambulatory monitoring data due to the limited control over the measurement procedures and elimination of unreliable measurements. The potential application area includes not only the analysis of biological data series but also extends to astronomical, climatological, hydrological, meteorological and other data series. The integrated solution providing the universal analysis algorithm for data sets from different sources is especially interesting for elicitation of various cross-modulation effects like governing of physiological rhythms by astronomic cycles.