Digital Avatars: An Application to eHealth

The development of digital health technologies allows, among other advantages, the remote monitoring of patients’ state of health. In this context, smartphones have become Telemedicine tools, as they have a high computational capacity and are equipped with sensors that make them aware of their environment and allow remote interaction. However, as in other ubiquitous computing systems, the storage, processing, and exchange of large amounts of data could put privacy at risk. In this paper, we present Digital Avatars, a software framework that allows an intensive use of the computational capabilities of smartphones following the People as a Service paradigm. This framework allows the development of evolving sociological profiles that guarantee the user’s personal confidentiality without considering external servers, and minimizing the exchange of information. To demonstrate the correct functioning of Digital Avatars in the field of health, we have developed and successfully validated a use case for the monitoring of patients suffering from Parkinson’s disease. The implemented system is composed of two applications (for smartphone and smartwatch, respectively), using Complex Event Processing, Fast Fourier Transform and Recurrent Neural Networks.