Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü, Türkiye
Tezin Onay Tarihi: 2015
Öğrenci: ANIL PAÇACI
Danışman: İSMAİL SENGÖR ALTINGÖVDE
Özet:One of the important aspects of the clinical research studies carried out in the pharmacovigilance and pharmacoepidemiology is the postmarketing drug surveillance. Utilization of the available Electronic Health Record (EHR) data is key to complement and strengthen the postmarketing safety studies. In addition, EHRs provide a huge, but still under-utilized source of information for the observational studies in clinical research. However, use of different EHR data models and vocabularies presents an important interoperability challenge. Predominant solution to this problem is to transform the data from these disparate EHR datasets into a common data model (CDM) in order to enable large-scale systematic analyses. Existing transformation practices depend on proprietarily developed Extract - Transform - Load (ETL) procedures. It requires a significant amount of expertise in both source and target models, as well as detailed technical knowledge on the underlying database implementations. Moreover, the experience gained during the transformation of one source is not readily transferable to other domains. In this thesis, we address these challenges and develop the necessary semantic transformation machinery to translate the EHR data available in SALUS Common Information Model to the Observational Medical Outcomes Partnership (OMOP) CDM. It enables pharmacovigilance researchers to seamlessly run existing safety analysis methods defined in the OMOP project on top of disparate EHR sources. The semantic materialization technique is adopted with the use of semantic mapping rules for data conversion on EYE reasoner. Accuracy and feasibility of the proposed framework have been evaluated in real-world settings together with pharmacovigilance researchers.