Sepski: Data-driven AI systems for individualized early detection and treatment optimization of sepsis in hospitals
- Duration: 01.01.2022 - 01.03.2023
Funding Organization: Ministerium für Wissenschaft und Gesundheit, RlP
Prof. Dr. Maik Kschischo
Koblenz University of Applied SciencesSepsis, colloquially called blood poisoning, is a life-threatening complication of infectious diseases, including severe COVID-19 infections, bacterial infections or viral flu. With approximately 75,000 deaths per year, sepsis is the third leading cause of death in Germany. However, there are also dramatic health consequences for survivors. The severity of the consequential damage and the chances of survival increase with early diagnosis and a fast and targeted therapy. There are also dramatic health consequences for survivors. The severity of the consequential damage and the chances of survival improve with early diagnosis and a fast and targeted therapy. However, early diagnosis is often difficult and medical decisions about optimal therapies are not always easy to make. In the project, artificial intelligence (AI) methods are used to
In the project, artificial intelligence (AI) methods are devised to observe patients automatically and to to detect dangerous patients earlier. In addition, we will design methods to infer individualized therapeutic strategies from data. This should be a first step towards a computer based decision support system, which supports doctors with their decision making about therapies with potentially detrimental side effects.
AI Focus Areas of the Research Project
Basic Research
- Machine Learning (ML): (Semi) Supervised Learning, Bayesian Neural Networks, Model Based, Generative Models
- Knowledge-Based Systems: Causality