Through the Medical Informatics Initiative (MII) and the establishment of the Data Integration Centres (DIC), clinical care data from various sources of Hospital Information Systems (HIS) are made available for medical research. With the methodological use case Phenotyping Pipeline (PheP), the SMITH consortium supports the development, qualitative enrichment and evaluation of the data. The University of Leipzig leads the project.

New approaches in scientific research

Server - Copyright: Liljam/shutterstock.com

PheP is a platform that enables clinical researchers to work together with statisticians and computer scientists in interdisciplinary collaboration to pursue scientific issues that previously seemed economically and technologically unthinkable. New scientific analyses are systematically conceived. Calculations on the existing data lead to new patient-related information. These determinable characteristics of the patients, so-called phenotypes, enrich the data set and are subsequently made available for further applications. Clinical research and patient care can be thereby optimized in the long term.

Reliable technologies for sensitive patient data

The technical prerequisite is provided by the PheP-Engine - a software solution built at each of our locations. This enables the introduction of an automatic and safe technology for the distribution and execution of algorithms. This allows sensitive patient data to remain in the clinic. The transfer and standardization of the data takes place via standardized interfaces.

Natural Language Processing (NLP) is a subproject of PheP. Certainly, not all data in electronic medical records is already available as uniquely coded data elements. Diagnoses, results, medication, side effects or laboratory data are extracted from the documents existing in the Hospital Information System (HIS) using methods of natural language processing and semantic text analysis.

Zitat Prof. Löffler PHEP

Sustainable improvement of patient care through research

The clinical use cases ASIC (Algorithmic Monitoring in Intensive Care) and HELP (Target-oriented antibiotic therapy in infection medicine) of the SMITH consortium exemplify the new possibilities. The methodological use case PheP is now establishing processes and an infrastructure to enable cross-site collaboration to answer future research questions.

The PheP concept also forms the basis of the cross-MII use case POLAR - Polypharmacy, Drug Interactions and Risks of MII, which was launched in early 2020 and involves all four consortia.