Medizinische Informatik, Statistik und Dokumentation

Research area semantics and ontologies in medicine

PI: Stefan Schulz

Focus: The focus of the group is on semantic data modelling of data from science and clinical practice. Two paths are pursued: the construction of symbolic knowledge by experts on the one hand, and automatic knowledge acquisition using machine learning on the other hand. In the former case, data are standardized by terminologies, ontologies and information models, in the latter case semantics is expressed by probabilistic and neural models. Most of the data is only available as text, which explains the focus set on text mining methods. The standardized data extracts obtained support document research, data analysis and clinical decision-making.

Networking: Currently important cooperation partners of the team are KAGes and CBmed, Graz, ELGA GmbH Vienna, as well as Roche Diagnostics (Basel and Belmont). In Germany, there are close contacts with the text mining company Averbis GmbH, Freiburg, EMPIRICA GmbH, Bonn, the Universities of FreiburgandTU Munich, the University of Jena, the Charité Berlin and DFKI Saarbrücken. Global contacts exist through active participation in the standardization organizations SNOMED International and HL-7. Current collaborations with colleagues from the universities of Trondheim, Murcia, Bordeaux, Ljubljana, Buffalo, the Bern University of Applied Sciences and the PUCPR (Brazil) should also be emphasized.


Digital Biomarkers

  • We are coordinators of the project DBM4PM (Digital Biomarkers for Precision Medicine) via CBmed. It aims at data extraction from clinical documents through semantic standardization of data for the tumour board platform NAVIFY®. The analysis of large amounts of clinical bulk data is another focus. They consist in brief diagnosis descriptions, routinely annotated with ICD-10 codes. The aim is to increase data quality and to acquire clinical terminology by using machine learning. Finally, we provide support to a project that predicts the risk of clinical complications (delirium) based on KAGes data.
  • Duration: 2015-2023 
  • Funded by: FFG
  • Project partners: Roche Diagnostics; CBmed; Averbis GmbH; KAGes

Localisation of SNOMED CT

  • The international terminology standard SNOMED CT is now available in all German-speaking countries, but it has not yet been translated into German. We have been creating a collection of German clinical terms linked to SNOMED CT codes, using both top-down and bottom-up approaches. The result, the "Graz (German) Interface Terminology for SNOMED CT (GIT-SCT)" is used experimentally in own projects as well as by cooperation partners for text mining.
  • Duration: 2015-2022
  • Funded by: Med Uni Graz
  • Project partners: Averbis GmbH, Freiburg, Germany; ELGA GmbH, Vienna

PRECISE4Q - Stroke Prediction Models

  • The Horizon 2020 project Precise4Q aims to create data-driven stroke prediction models in four phases. While the models incorporate new deep learning approaches, data normalization and data integration are clear bottlenecks. Data comes from several countries, different suppliers, in highly different structures. Our task is semantic harmonization. Currently the focus is on the manual annotations of clinical texts as training material for models for information extraction.
  • Duration: 2018-2022
  • Funded by: European Commission
  • Project partners: empirica GmbH, Bonn; TU Dublin; University College Dublin; Health Ethics and Policy Lab, University of Zurich; Guttmann Institute, Barcelona; University of Linköping; DFKI Saarbrücken; AOK NO - GeWINO, Berlin; QMENTA, Barcelona; University of Murcia.

Principal Investigator

Stefan Schulz 
T: +43 316 385 16939