The main objective of our research is to translate health data into better treatment decisions through the use of data analytics. For this purpose, we develop innovative methodology from the fields of statistics, machine learning and big data in order to contribute to further research developments in data-driven decision support across various tasks in health management. Our work has been published in leading journals in the fields of Information Systems and Operations Research, as well as computer science conferences.
Data-Driven Allocation of Preventive Care
Healthcare is increasingly focusing on preventive treatments to improve disease progression or even prevent diseases altogether. Unfortunately, preventive treatment of large populations is not feasible due to cost constraints, which raises the question in healthcare management of which patients benefit most from treatments. In this project, we develop a decision model based on machine learning and optimization, which finds the subgroup of patients who benefit most from treatment for a fixed budget for preventive treatments. Accordingly, the expected number of diseases is minimized.
A Hypoglycaemia Warning System for the Prevention of Road Traffic Accidents
Hypoglycemia has consistently been shown to be associated with an increased risk of driving mishaps. As a prevention, this project investigates if machine learning models can be used to detect hypoglycemia in an early stage to trigger direct interventions through a support module.
The project is led by Prof. Dr. Christoph Stettler from the Inselspital Bern / University of Bern and Prof. Dr. Elgar Fleisch from the Center for Digital Health Interventions, University of St. Gallen and ETH Zurich. Prof. Dr. Mathias Kraus contributes to the development of the machine learning module. Find the project page at https://www.dcberne.com/de/projekte/headwind/.