Research
The Section for Clinical Research IT (SKFiT) is dedicated to advancing clinical research IT infrastructure, aiming to generate medical value through the utilization of extended data sources. Our research focuses on the following key areas:
External Data
SKFiT explores and evaluates methods to harness external data formats—such as multimodal time series, video data, and patient-reported outcome measures (PROMs)—collected via smartphone apps, wearables, or ambient sensors in medical or everyday contexts.
Current Research Projects:
Medical Standards and Terminologies
Standards form the backbone of data exchange in healthcare. To ensure interoperability between systems, the adoption of expressive and easily implementable standards is essential. Engaging the developer community directly in the (further) development of standards allows for the incorporation of expert requirements from the source.
SKFiT actively contributes to the advancement of the international standard HL7 FHIR (Fast Healthcare Interoperability Resources). Medical coding systems enable the structured capture and seamless exchange of health data, such as from electronic health records (EHRs), facilitating interoperable communication and analysis.
To achieve this, coding systems like LOINC, SNOMED CT, and ICD-10 are employed. Utilizing standards like FHIR or openEHR necessitates proficient and comprehensive handling of terminologies to accurately represent the information being conveyed. Efficient access to these coding systems requires a terminology server that provides centralized, machine-readable resources.
SKFiT offers a solution that serves as a central platform within national initiatives like the Medical Informatics Initiative (MII) and the Network University Medicine (NUM), integrating all common coding systems and simplifying their usage.
Through the distributed infrastructure of MII and NUM, researchers can request access to the curated data for their projects. SKFiT plays a significant role in the ongoing development of the central Research Data Portal for Health (FDPG).
Current Research Projects:
-
Medical Data Integration Center (MeDIC) within the Medical Informatics Initiative and the Network University Medicine
Semantic Data Integration
Semantic data integration in healthcare is a crucial step toward harmonizing fragmented and heterogeneous health data from various sources and formats, making them accessible for effective use. This integration aims to connect different data, potentially structured in diverse coding systems, and present them in a unified, machine-readable format.
A common standard for this purpose is FHIR (Fast Healthcare Interoperability Resources). The terminology server provided by SKFiT is highly relevant in this context. During data integration, not only the data being transformed from one representation to another are important, but also the metadata describing the source and target data.
By utilizing a Metadata Repository within MeDIC, these metadata are employed during mapping to ensure traceable data integration. The resulting data are made available through the distributed infrastructure of MII and NUM, allowing researchers to link UKSH data with data from other sites.
SKFiT is actively involved in the further development of the central Research Data Portal for Health (FDPG).
Current Research Projects:
-
Medical Data Integration Center (MeDIC) within the Medical Informatics Initiative and the Network University Medicine
Data Analytics and Decision Support
SKFIT investigates and evaluates methods for medical data analysis and interpretation based on deep neural networks. These methods are applied to extract medically relevant interpreted parameters from multimodal data. For example, SKFIT processes multimodal time series, image and video data, as well as medical texts and EHRs.
Furthermore, SKFIT researches decision support methods (deep neural networks, case-based reasoning, rule-based approaches) to derive, for example, recommended actions based on this extended data foundation.
Current Research Projects:
Personalized Tele-Intervention
SKFiT investigates personalized closed-loop tele-intervention approaches — for example, in the context of video-supported therapy sessions with online feedback on execution quality focusing on tailoring remote healthcare interventions to individual patient needs.. Corresponding apps are also being designed and implemented iteratively with the involvement of the user groups concerned, and tested with them for usability.
Current Research Projects:
Clinical Research Infrastructure and Data Protection
The development of a generic data protection and consent concept, the implementation of semantic data integration using structured documentation and standards, the implementation of a metadata repository, the validation of open-source software for research data management, the development of minimal data sets, and the establishment of data warehouse solutions.
Due to its central role within the Research Data Portal for Health (FDPG), SKFIT is also active in this area within the national infrastructure of the Medical Informatics Initiative and the Network University Medicine.
Current Research Projects:
