The “Big Data in Health” Conference was held at the Villa Umbra, in Pila (Perugia, Italy), June 29-30, 2017. The event was sponsored by the University of Perugia and the University of Salerno, and organized by Enrico Capobianco (IDSC Computational Biology & Bioinformatics), Luca Pieroni and Luca Salmasi (Department of Economics, University of Perugia, Italy), Liliana Minelli (Department of Medicine, University of Perugia, Italy), Sergio Pagano and Pierpaolo Cavallo (Departments of Physics and Department of Medicine, University of Salerno, Italy).

About 60 people registered and participated to the 5 sessions in the program. The audience was a mix of clinicians, academicians, researchers in industry, health administrative sector-, and other health professionals, thus confirming the spirit of the initiative aimed to propose an integrative multidisciplinary approach to Big Data in Health, particularly from the Electronic Health Records perspective.

The first session was dedicated to Institutional contributions, in particular Regione Umbria (which co-sponsored the meeting), and the GARR Director Dr Federico Ruggieri (who illustrated the infrastructure needed for the management of Big Data in large projects). Three other sessions followed with contributions from academics and research, and including four young scholars who presented the results of their studies (PhD theses and postdoc projects). Finally, a well-participated panel session concluded the meeting.

The main established principle characterizing the workshop was fully reflected by the fact that knowledge is dependent not just on the quantity of available Big Data, but on their connectivity. It is therefore the network between people and ideas underlying the generation, analysis, and validation of such data that creates the premises for fruitful synergies. Consequently, integrative methods can be designed to infer about the correlative and causative relationships between the complex health- and disease-related variables. Note that often Big Data do not directly reveal associations and dependencies, thus they need to be processed by sophisticated computational approaches.

One emerging aspect is the fact that the integrability between clinical and administrative data in support to both research and clinical decision making is destined to play a crucial role. Once this is reached, interoperability and actionability are required steps. These two main data types are generated in response to different questions, first of all, which explains some current bottlenecks. Examples were brought to the general attention, involving diagnostic imaging in cancer, pharmacosurveillance, omics and microbiomics, among others. Finally, interesting applications were presented in the fields of cardiology, nephrology, aging, among others, and, interestingly, appeared the relevance of concepts like “inflammaging”, “immunobiography”, and “digital biomarkers”.