Proposal of an Architecture to support High Quality Automatic Data Collection in the context of Multi-Centric Studies

The increasing ease of people to move from one place to another and the rapid emergence of multi-centric clinical trials make necessary an extensive multilayer integration of data from different health areas so that it is possible to minimize the need for human intervention. The idea behind this work is to exploit the information contextualization properties guaranteed by the Clinical Document Architecture Release 2.0 (CDA R2) together with the skills of Machine Learning so that it is possible to highlight values out of the therapeutic range or outside the range which generally data belong to. The proposed architecture it is composed by three elements designed for the purpose of supporting the automatic transfer of high-quality data from one system to another and to point out any outliers. The architecture supports the creation of a large, well-structured and well-contextualized database for multi-centric clinical studies.