The cloud4health project: Secondary use of clinical data with secure cloud-based text mining services

Abstract

Advances in translational and personalized medicine require the integration of multiple patient related resources across different organizational bodies. Thus, secure cloud environments for huge data processing, storage and data integration are needed. Moreover, the integration of clinical patient data is indispensable for translational research. Although operational e-health record systems are established in most hospitals, many clinical and phenotypically relevant parameters can only be found in unstructured texts like medical records and reports. To meet these challenges, the cloud4health project established a cloud-based text mining platform to facilitate information extraction of biomedical texts in a secure cloud environment. In order to comply with privacy regulations, general technical demands and security rules for such a cloud installation were developed and have been implemented. Different clinical use cases show the wide spectrum of application of specific text mining services in a secure cloud environment. As application examples, two use cases utilizing text mining technologies to analyse pathology and surgery reports are analysed in detail.