Design and Evaluation of an AI-Powered Clinical Decision Support System for Early Diagnosis of Sepsis in Intensive Care Units
Keywords:
Sepsis, ICU, Clinical Decision Support System, Artificial Intelligence, Early Diagnosis, Machine Learning, MIMIC-IIIAbstract
Sepsis is a major cause of mortality in intensive care units (ICUs), and early diagnosis significantly improves survival outcomes. Traditional scoring systems such as SIRS and SOFA are limited in sensitivity and timeliness. This paper presents the design and performance evaluation of an Artificial Intelligence-powered Clinical Decision Support System (AI-CDSS) aimed at the early diagnosis of sepsis using ICU patient data. Utilizing supervised machine learning models and retrospective datasets, the system was trained and validated for accuracy, sensitivity, and lead-time performance.
Focusing on data from before 2022, particularly the widely used MIMIC-III database, this study evaluates the effectiveness of random forest and logistic regression models for sepsis prediction. The AI-CDSS demonstrated a lead time of 5.6 hours before clinical recognition, with an AUROC exceeding 0.85. The paper also reviews literature prior to 2022 and highlights how AI-driven CDSS can be better integrated into critical care workflows through interpretability and usability enhancements
References
Desautels, T., Calvert, J., Hoffman, J., et al. (2016). Prediction of sepsis in the intensive care unit with minimal electronic health record data: A machine learning approach. JMIR Medical Informatics, 4(3), e28.
Johnson, A. E. W., Pollard, T. J., Shen, L., et al. (2016). MIMIC-III, a freely accessible critical care database. Scientific Data, 3, 160035.
Nemati, S., Holder, A., Razmi, F., et al. (2018). Interpretable machine learning models for the prediction of sepsis in ICU. Critical Care Medicine, 46(4), 547–553.
Henry, K. E., et al. (2015). A targeted real-time early warning score (TREWScore) for septic shock. Science Translational Medicine, 7(299), 299ra122.
Kam, H. J., & Kim, H. Y. (2017). Learning representations for the early detection of sepsis with deep neural networks. Computers in Biology and Medicine, 89, 248–255.
Fleuren, L. M., Klausch, T. L. T., Zwager, C. L., Schoonmade, L. J., Guo, T., Roggeveen, L. F., & Thoral, P. (2020). Machine learning for the prediction of sepsis: A systematic review and meta-analysis of diagnostic test accuracy. Intensive Care Medicine, 46(3), 383–400.
Giannini, H. M., Ginestra, J. C., Chivers, C., Draugelis, M. E., Hanish, A., Schweickert, W. D., & Fuchs, B. D. (2019). A machine learning algorithm to predict severe sepsis and septic shock: Development, implementation, and impact on clinical practice. Critical Care Medicine, 47(11), 1485–1492.
Taylor, R. A., Pare, J. R., Venkatesh, A. K., Mowafi, H., Melnick, E. R., Fleischman, W., & Hall, M. K. (2016). Prediction of in-hospital mortality in emergency department patients with sepsis using machine learning. Academic Emergency Medicine, 23(3), 269–278.
Komorowski, M., Celi, L. A., Badawi, O., Gordon, A. C., & Faisal, A. A. (2018). The Artificial Intelligence Clinician learns optimal treatment strategies for sepsis in intensive care. Nature Medicine, 24(11), 1716–1720.
Kam, H. J., & Kim, H. Y. (2017). Learning representations for the early detection of sepsis with deep neural networks. Computers in Biology and Medicine, 89, 248–255.
Moor, M., Horn, M., Rieck, B., Schneider, G., & Borgwardt, K. (2021). Early prediction of sepsis in the ICU using machine learning: A systematic review. NPJ Digital Medicine, 4(1), 1–13.
Gultepe, E., Green, J. P., Nguyen, H., Adams, J., Albertson, T., & Tagkopoulos, I. (2014). From vital signs to clinical outcomes for patients with sepsis: a machine learning basis for a clinical decision support system. Journal of the American Medical Informatics Association, 21(2), 315–325.
Barton, C., Hu, L., Choi, E., & Davis, M. (2019). Evaluation of sepsis screening tools in the emergency department: A comparative retrospective analysis. BMJ Open, 9(1), e022341.
Reyna, M. A., Josef, C., Jeter, R., Shashikumar, S. P., Westover, M. B., Nemati, S., & Clifford, G. D. (2020). Early prediction of sepsis from clinical data: the PhysioNet/Computing in Cardiology Challenge 2019. Critical Care Medicine, 48(2), 210–217.
Shashikumar, S. P., Stanley, M. D., Sadiq, I., Li, Q., Holder, A., Clifford, G. D., & Nemati, S. (2017). Early sepsis detection in critical care patients using multiscale blood pressure and heart rate dynamics. Journal of Electrocardiology, 50(6), 739–743
