HEALTH DATA
Data driven methods are key factors for solving research questions in the context of health of humans as well as animals. BIN researchers develop data science and data engineering systems that analyze data from the intramural as well as extramural treatments of patients as well as data from life stock and food production.
In the research area Health Data, we focus on the development and application of data science and engineering systems dedicated to the improvement of the health of humans as well as animals. For instance, we develop information systems that aggregate, process, and analyze medical data from patients in hospitals. The LeiVMed system is developed for the Oberösterreichische Gesundheitsholding in close collaboration with our colleagues at Campus Steyr; this system is already in use in several hospitals in Upper Austria and significantly helps increasing the quality of treatment of patients. Within the doctoral program PLFDoc we conduct research in precision life stock farming together with colleagues at Veterinarian Medical University Vienna and the Technical University Vienna; we also joined the FFOQSI consortium in which we analyze data from food production and food processing. Furthermore, we join forces with ELGA and other institutions in the standardization of medical data handling. One of the newest research developments is our engagement in the support of caregivers, especially for seniors, where we concentrate on dementia research in close cooperation with our partner MAS Alzheimerhilfe.
In close cooperation with our partners from research and industry, we develop algorithms for preprocessing and analyzing data using statistical methods and machine learning, as well as visualizing and interpreting the results. In many of our projects, we see the need for customized data processing pipelines due to the heterogeneity and complexity of data structures in real world processes and systems. We use applied statistics as well as numerous machine learning approaches, including black box methods (deep neural nets, random forests, gradient boosted trees, etc.) and white box methods (symbolic regression by genetic programming). We build our knowledge discovery pipelines on a variety of different frameworks, especially python scikit-learn, tensorflow, pytorch, MATLAB, and HeuristicLab.
