60-hours Sepsis Prediction in ICU Patients

• Alícia Marzola Chaves UFMG
• João M. F. Santos UFMG
• Letícia Ribeiro Miranda UFMG
• Alexandre G. Barros UFMG
• Wagner Meira Jr. UFMG
• Gisele Pappa UFMG

DOI: https://doi.org/10.5753/eniac.2024.245285

Abstract

This study explores the use of simple machine learning techniques to predict the development of sepsis in patients admitted to the intensive care unit (ICU). Using data from the interval of 6 hours before ICU admission to 6 hours after admission, analyzing 26 attributes of patients who met the criteria for predicting sepsis and balancing the number of septic and non-septic cases to prevent the model from becoming biased, the implemented methods were able to predict people who acquired sepsis in the first 60 hours after the aforementioned period. Despite the high number of missing data and the high rate of false negatives, the results of these analyses are important for the implementation of preventive care and for the possibilities of continuous improvement of the model.

Keywords: Artificial Intelligence, Sepsis, ICU

References

Ahmed, A., Liu, J., & Zhang, Z. (2023). Improving the Timeliness of Early Prediction Models for Sepsis through Utility Optimization. IEEE.

Cunha, B. A. (2008). Sepsis and septic shock: selection of empiric antimicrobial therapy. Critical Care Clinics, 24(2), 313-334, ix.

eICU Collaborative Research Database. Disponível em: [link]. Acesso em: 28 agosto de 2024.

Fleischmann, C., Scherag, A., Adhikari, N. K. J., Hartog, C. S., Tsaganos, T., Schlattmann, P., ... & Reinhart, K. (2016). Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. American Journal of Respiratory and Critical Care Medicine, 193(3), 259-272.

Henry, K. E., Hager, D. N., Pronovost, P. J., & Saria, S. (2015). A targeted real-time early warning score (TREWScore) for septic shock. Science Translational Medicine, 7(299), 299ra122.

Kieft, H., Hoepelman, A. I. M., Zhou, W., Rozenberg-Arska, M., Struyvenberg, A., & Verhoef, J. (1993). The sepsis syndrome in a Dutch University Hospital. Clinical observations. Archives of Internal Medicine.

Kumar, A., Roberts, D., Wood, K. E., Light, B., Parrillo, J. E., Sharma, S., ... & Cheang, M. (2006). Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical Care Medicine, 34(6), 1589-1596.

Liu, C., Feng, S., & Luo, J. (2022). Sepsis Detection using Neural Networks. IEEE.

Liu, V. X., Fielding-Singh, V., Greene, J. D., Baker, J. M., Iwashyna, T. J., Bhattacharya, J., & Escobar, G. J. (2017). The Timing of Early Antibiotics and Hospital Mortality in Sepsis. American Journal of Respiratory and Critical Care Medicine, 196(7), 856-863.

Morley, P., Wang, Y., & Zhao, Z. (2021). Exploring Features Contributing to the Early Prediction of Sepsis Using Machine Learning. IEEE.

Nahar, A., Hemanth, K. S., & Abirami, G. (2020). Early Detection of Sepsis Using Feature Selection, Feature Extraction, and Neural Network Classification. IEEE.

Singer, M., Deutschman, C. S., Seymour, C. W., Shankar-Hari, M., Annane, D., Bauer, M., ... & Angus, D. C. (2016). The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 315(8), 801-810.

Wang, J., Li, Q., & Yuan, H. (2023). Early Prediction of Sepsis using Time Series Forecasting. IEEE.

World Health Organization. Guidelines on the Clinical Management of Sepsis. Disponível em: [link]. Acesso em: 28 de agosto de 2024.

Zhang, X., & Chen, Y. (2022). A Xgboost Based Algorithm For Early Prediction Of Human Sepsis. IEEE.