11 / 2020 – 05 / 2021
Sepsis is a common and deadly disorder leading to more deaths per year than breast cancer and prostate cancer combined . While not receiving any antibiotic treatment, the mortality rate drastically increases by 7.6% per hour . An early diagnosis therefore has a high impact on
patient outcome and is a key factor in sepsis treatment [1, 3]. A solution to help physicians with finding a diagnosis by using various parameters are clinical decision support systems. Early sepsis prediction algorithms are an example for such systems. However, they have not been used in clinical practice yet, due to the wide range of symptoms sepsis can yield and no unique biomarker that can diagnose sepsis confidently until now [3, 4]. This leads to algorithms that are not specific and sensitive enough for clinicians to be useful [4, 5].
The assumption though is, that sepsis diagnosis depends on local conditions like the definition of sepsis, available biomarkers, or demographics . Therefore, an algorithm might be greatly improved by incorporating local circumstances into the development process.
Thus, the aim of this research is to investigate how local circumstances differ in various countries and settings and to validate two algorithms with both publicly available patient data merged from several ICU departments and local patient data.
 Ron Daniels. Surviving the first hours in sepsis: getting the basics right (an intensivist’s perspective). Journal of antimicrobial chemotherapy, 66(suppl_2):ii11ii23, 2011.
 Anand Kumar, Daniel Roberts, Kenneth E Wood, Bruce Light, Joseph E Parrillo, Satendra Sharma, Robert Suppes, Daniel Feinstein, Sergio Zanotti, Leo Taiberg, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical care medicine, 34(6):15891596, 2006.
 Yong Liu, Jun-huan Hou, Qing Li, Kui-jun Chen, Shu-Nan Wang, and Jian-min Wang. Biomarkers for diagnosis of sepsis in patients with systemic inflammatory response syndrome: a systematic review and meta-analysis. Springerplus, 5(1):2091, 2016.
 Michael J Pettinati, Gengbo Chen, Kuldeep Singh Rajput, and Nandakumar Selvaraj. Practical machine learning-based sepsis prediction. In 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pages 49864991. IEEE, 2020.
 Nathan Peier-Smadja, Timothy Miles Rawson, Raheelah Ahmad, Albert Buchard, Georgiou Pantelis, F-X Lescure, Gabriel Birgand, and Alison Helen Holmes. Machine learning for clinical decision support in infectious diseases: a narrative review of current applications. Clinical Microbiology and Infection, 2019.
 Greg S Martin, David M Mannino, Stephanie Eaton, and Marc Moss. The epidemiology of sepsis in the united states from 1979 through 2000. New England Journal of Medicine, 348(16):15461554, 2003.