A Big Data Architecture for the Extraction and Analysis of EHR Data

In the current Italian eHealth scenario, a national IT platform has been designed and developed with the purpose of ensuring the interoperability between the various Electronic Health Record (EHR) systems that have been adopted in the different regions of the country, according to the requirements provided by Italian Laws. In this way, the healthcare providers and the policy makers can acquire and process the data of a patient despite its initial format and source, allowing an improved quality of patient care and optimizing the management of the financial resources. To further exploit this huge resource of health and social data, it is very important to allow the extraction of the complex information buried under the Big Data source enabled by the EHRs, providing the physicians, the researchers and public health policy makers with innovative instruments. Meeting this need is not a trivial task, due to the difficulties of processing different document formats and processing Natural Language text, alongside to the problems related to the data size. In this paper we propose a Big Data architecture that is able to extract information from the documents acquired by the EHRs, integrate and process them, providing a set of valuable data for both physicians and patients, as well as decision makers.