Comparison of different AI systems for diagnosing sepsis, septic shock, and cardiogenic shock: a retrospective study

被引:0
作者
Obmann, Dirk [1 ,2 ]
Muench, Philipp [3 ]
Graf, Bernhard [2 ]
von Jouanne-Diedrich, Holger [3 ]
Zausig, York A. [1 ,2 ]
机构
[1] Klinikum Aschaffenburg Alzenau, Dept Anaesthesiol & Crit Care, Aschaffenburg, Germany
[2] Univ Regensburg, Dept Anaesthesiol, Regensburg, Germany
[3] TH Aschaffenburg Univ Appl Sci, Competence Ctr Artificial Intelligence, Fac Engn, Aschaffenburg, Germany
来源
SCIENTIFIC REPORTS | 2025年 / 15卷 / 01期
关键词
Artificial intelligence; Ai; Decision support systems; Sepsis; Septic shock; Cardiogenic shock;
D O I
10.1038/s41598-025-00830-9
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
Sepsis, septic shock, and cardiogenic shock are life-threatening conditions associated with high mortality rates, but differentiating them is complex because they share certain symptoms. Using the Medical Information Mart for Intensive Care (MIMIC)-III database and artificial intelligence (AI), we aimed to increase diagnostic precision, focusing on Bayesian network classifiers (BNCs) and comparing them with other AI methods. Data from 5970 adults, including 950 patients with cardiogenic shock, 1946 patients with septic shock, and 3074 patients with sepsis, were extracted for this study. Of the original 51 variables included in the data records, 12 were selected for constructing the predictive model. The data were divided into training and validation sets at an 80:20 ratio, and the performance of the BNCs was evaluated and compared with that of other AI models, such as the one rule classifier (OneR), classification and regression tree (CART), and an artificial neural network (ANN), in terms of accuracy, sensitivity, specificity, precision, and F1-score. The BNCs exhibited an accuracy of 87.6% to 91.5%. The CART model demonstrated a notable 91.6% accuracy when only three decision levels were used, whereas the intricate ANN model reached 90.5% accuracy. Both the BNCs and the CART model allowed clear interpretation of the predictions. BNCs have the potential to be valuable tools in diagnostic tasks, with an accuracy, sensitivity, and precision comparable, in some cases, to those of ANNs while demonstrating superior interpretability.
引用
收藏
页数:10
相关论文
共 39 条
[1]  
Ahmed M., 2020, Annals of Data Science, V7, P427, DOI DOI 10.1007/S40745-019-00237-0
[2]  
Atkinson Beth, 2023, CRAN
[3]   Development of a machine learning model to predict the risk of late cardiogenic shock in patients with ST-segment elevation myocardial infarction [J].
Bai, Zhixun ;
Hu, Shan ;
Wang, Yan ;
Deng, Wenwen ;
Gu, Ning ;
Zhao, Ranzun ;
Zhang, Wei ;
Ma, Yi ;
Wang, Zhenglong ;
Liu, Zhijiang ;
Shen, Changyin ;
Shi, Bei .
ANNALS OF TRANSLATIONAL MEDICINE, 2021, 9 (14)
[4]   Incidence and Outcomes of Nontraumatic Shock in Adults Using Emergency Medical Services in Victoria, Australia [J].
Bloom, Jason E. ;
Andrew, Emily ;
Dawson, Luke P. ;
Nehme, Ziad ;
Stephenson, Michael ;
Anderson, David ;
Fernando, Himawan ;
Noaman, Samer ;
Cox, Shelley ;
Milne, Catherine ;
Chan, William ;
Kaye, David M. ;
Smith, Karen ;
Stub, Dion .
JAMA NETWORK OPEN, 2022, 5 (01) :E2145179
[5]  
Breiman L, 1998, WADSWORTH INT GROUP, DOI DOI 10.1785/0120150058
[6]   Septic Cardiomyopathy: From Pathophysiology to the Clinical Setting [J].
Carbone, Federico ;
Liberale, Luca ;
Preda, Alberto ;
Schindler, Thomas Hellmut ;
Montecucco, Fabrizio .
CELLS, 2022, 11 (18)
[7]   Early Prediction of Cardiogenic Shock Using Machine Learning [J].
Chang, Yale ;
Antonescu, Corneliu ;
Ravindranath, Shreyas ;
Dong, Junzi ;
Lu, Mingyu ;
Vicario, Francesco ;
Wondrely, Lisa ;
Thompson, Pam ;
Swearingen, Dennis ;
Acharya, Deepak .
FRONTIERS IN CARDIOVASCULAR MEDICINE, 2022, 9
[8]  
Davenport Thomas, 2019, Future Healthc J, V6, P94, DOI 10.7861/futurehosp.6-2-94
[9]   Comorbidity measures for use with administrative data [J].
Elixhauser, A ;
Steiner, C ;
Harris, DR ;
Coffey, RN .
MEDICAL CARE, 1998, 36 (01) :8-27
[10]  
Evans L, 2021, INTENS CARE MED, V47, P1181, DOI [10.1007/s00134-021-06506-y, 10.1097/CCM.0000000000005337]