The amount of available Big data has grown drastically in the last decade, and a faster growth rate is expected in the coming years. Specifically, various biomedical domain methods, e.g., liquid biopsies, medical images, and genome sequencing, produce large volumes of data from where new biomarkers, or biological characteristics and medical signs, can uncover the incidence of a disease. Clinicians are faced with several challenges when analyzing biomedical data sources during diagnosis and treatment prescriptions. Biomedical data are presented in countless formats such as medical records, images, or genome data, that have to then be combined for optimal therapy decisions. Lastly, different regulation for enforcing data protection and privacy may hinder free access to biomedical data. This chapter addresses challenges present during the management of big biomedical data and presents a data-driven framework that resorts to ontologies to describe the main characteristics of data sources whose access is regulated by different data access regulations. The Privacy Ontology is defined as a formalism for representing the various entities that play a relevant role in the collection, anonymisation, integration, processing, and distribution of big biomedical data. As proof of concept, we illustrate the expressiveness of the proposed approach in the context of the European Union funded project iASiS, which aims at transforming big data into actionable knowledge to pave the way for personalised medicine and individualised treatments.
Paper