Digital Health Data, a Way to Take Under Control the Quality During the Elaboration Processes

In this work we explore the uncertainty in measured attributes of events, correlating with other events happening in a near time window. The conceptual framework of the method is the event graph concept. Our objective is to filter out the unreliable events and obtain an optimized fraction of data, in order to compute decent statistics, to be able to evaluate the data and build derived measures. Developing our framework, we proceed with some experiments using data from wearable devices, both synthetic data sets and a public available benchmark.