Skip to main content Skip to page footer

LeiVMed

Researchers

Louise Buur MSc.

Marina Strobl MSc.

Dr. Julia Vetter MSc.

Manuel Wechselberger BSc.

Prof.(FH) PD DI Dr. Stephan Winkler


Duration

Duration: 2017 – present

Research Areas

Health Data

Partners

University of Applied Sciences Upper Austria, Hagenberg Campus, RG Bioinformatics

University of Applied Sciences Upper Austria, Steyr Campus, Process Management and Business Intelligence

Oberösterreichische Gesundheitsholding GmbH (OÖG)

Kepler University Hospital (KUK)

Since 2015, FH Upper Austria and Oberösterreichische Gesundheitsholding (OÖG) have joined forces and develop the platform LeiVMed (LeistungsVergleich Medizin) for the monitoring and analysis of data of treatments of patients. Thus, LeiVMed is a benchmarking tool that enables participating hospitals to compare their clinical processes. It provides doctors, nursing staff, and higher-level management with a quick, up-to-date, and systematic overview. The center of the chosen representation is a “cockpit”, which puts medical and business decisions into context. This means that treatment and organizational processes are presented transparently, and managers are provided with the appropriate control options. As an independent benchmarking platform beyond company boundaries, the own performance can be compared with system partners. The transparent internal and external representations provide the basis for optimized and cost-efficient patient care: Quality increases and costs decrease. All data is collected from electronic documentation systems, transformed, and individually displayed on the LeiVMed web platform.

Currently, LeiVMed comprises data from more than 50 000 patient records. This includes, besides demographic information, the number of services, costs, and different risk factors. Since hospital costs continue to rise, further efforts to reduce costs seem inevitable. A crucial role plays the determination of complications in the clinical setting. Here, data science algorithms and workflows are developed for improved complication identification as well as risk analyses. Further, machine learning models are trained for improved bed management by predicting the length of stay after surgeries.