A Novel Approach for Clinical Data Harmonization

A city transforms into a “smart” one when it utilizes the data from smart devices, cloud infrastructures, applications and repositories in order to develop and provide new services, products and insights with the goal being to offer a safer, more efficient and robust environment for government, citizens and businesses, and accelerate sustainable economic growth. Although the variability across cities in terms of cultural background, demographics, current infrastructures, topology, among others, might drive significant variations in the individual goals of a smart city, the role of health in achieving the aforementioned goals is of paramount importance in any city. One of the starting points of introducing a health-related orientation into a smart city developments and operations is the intelligent processing of the overwhelming amount of clinical data being continuously generated during healthcare provision, daily activities and clinical research. Great challenges that need to be met in order to offer big data analytics over clinical data for the purposes of a smart city lie in their high heterogeneity at system, syntactic, structural and semantic level as well as their sensitive nature from the legal and ethical perspective, which may prevent and/or limit access to and analysis of the data. In this work we will present a novel approach for expressing clinical data using a common formalism with explicit semantics of the terms being used. Based on this established semantic ground, questions on health-related data can be placed for disease prevention, treatment monitoring, post-marketing surveillance, policy making, among others.

[1]  Ventola Cl Big Data and Pharmacovigilance: Data Mining for Adverse Drug Events and Interactions. , 2018 .

[2]  Cosmin Stroe,et al.  AgreementMaker: Efficient Matching for Large Real-World Schemas and Ontologies , 2009, Proc. VLDB Endow..

[3]  Nadine Cullot,et al.  Database-to-Ontology Mapping Generation for Semantic Interoperability , 2007 .

[4]  Boris Motik,et al.  OWL 2: The next step for OWL , 2008, J. Web Semant..

[5]  Erhard Rahm,et al.  Schema and ontology matching with COMA++ , 2005, SIGMOD '05.

[6]  R. Kush,et al.  Global clinical data interchange standards are here! , 2007, Drug discovery today.

[7]  Werner Ceusters,et al.  HL7 RIM: An Incoherent Standard , 2006, MIE.

[8]  Dickson Lukose,et al.  Ontology Alignment - A Survey with Focus on Visually Supported Semi-Automatic Techniques , 2010, Future Internet.

[9]  Jérôme David,et al.  Detection and Transformation of Ontology Patterns , 2009, IC3K.

[10]  Suchi Saria,et al.  Why policymakers should care about “big data” in healthcare , 2018, Health Policy and Technology.

[11]  Theodora A. Varvarigou,et al.  A novel semantic representation for eligibility criteria in clinical trials , 2017, J. Biomed. Informatics.

[12]  C. L. Ventola Big Data and Pharmacovigilance: Data Mining for Adverse Drug Events and Interactions. , 2018, P & T : a peer-reviewed journal for formulary management.

[13]  Jérôme Euzenat,et al.  Ontology Matching: State of the Art and Future Challenges , 2013, IEEE Transactions on Knowledge and Data Engineering.

[14]  Eric Miller,et al.  An Introduction to the Resource Description Framework , 1998, D Lib Mag..

[15]  Prashant Doshi,et al.  OPTIMA: Tool for Ontology Alignment with Application to Semantic Reconciliation of Sensor Metadata for Publication in SensorMap , 2008, 2008 IEEE International Conference on Semantic Computing.

[16]  Theodora A. Varvarigou,et al.  An Intelligent Ontology Alignment Tool Dealing with Complicated Mismatches , 2014, SWAT4LS.