AI-assisted sepsis detection: a systematic, meta-narrative review
Main Article Content
Keywords
Sepsis, Artificial Intelligence, Machine Learning, Computer-Assisted Diagnosis, Intensive Care Units, Sensitivity and Specificity
Abstract
Despite a growing primary literature, systematic reviews on the contribution of artificial intelligence (AI) to sepsis detection remain limited. Algorithmic models may enable earlier identification and save lives.
Objective: To synthesize evidence on AI performance for sepsis detection and its contribution to diagnostic accuracy.
Methods: Systematic review with a meta-narrative approach. Searches in PubMed, SciELO, and ScienceDirect included Spanish/English studies with ≥250 patients applying AI to sepsis detection/diagnosis. From 713 records, we extracted design, setting, algorithms, metrics (AUC/ROC, sensitivity, specificity), and comparisons with conventional methods. Results: Sixteen studies met inclusion criteria, predominantly in ICU settings. The most frequent algorithms were XGBoost, Random Forest, and Support Vector Machine. Two methodological paradigms emerged: (i) prediction with machine-learning models trained on clinical/monitoring data; and (ii) hybrid approaches combining traditional tools (e.g., qSOFA/SIRS, clinical rules) with AI. Several series reported AUC >0.90, with higher sensitivity and specificity than conventional approaches, particularly for early detection. Heterogeneity persisted in sepsis definitions, data sources, and external validation.
Conclusions: AI yields meaningful gains in sensitivity and specificity versus traditional methods for sepsis detection, especially in ICUs. Its greatest value arises when combined with clinical assessment and established scores, supported by external validation and prospective studies to ensure clinical effectiveness, interpretability, and diagnostic/therapeutic safety.
References
2. Kaul V, Enslin S, Gross SA. History of artificial intelligence in medicine. Gastrointest Endosc [Internet]. 2020 Oct 1 [citado el 19 de mayo de 2025];92(4):807–12. Disponible en: https://www.sciencedirect.com/science/article/pii/S0016510720344667
3. Impacto de la inteligencia artificial en salud pública | AlfaPublicaciones [Internet]. [citado el 19 de mayo de 2025]. Disponible en: https://www.alfapublicaciones.com/index.php/alfapublicaciones/article/view/562
4. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, Kievlan DR, et al. Global, regional, and national sepsis incidence and mortality, 1990–2017: analysis for the Global Burden of Disease Study. Lancet [Internet]. 2020 Jan 18 [citado el 28 de mayo de 2025];395(10219):200–11. Disponible en: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(19)32989-7/fulltext
5. Islam MM, Nasrin T, Walther BA, Wu CC, Yang HC, Li YC. Prediction of sepsis patients using machine learning approach: a meta-analysis. Comput Methods Programs Biomed. 2019 Mar;170:1–9.
6. Kijpaisalratana N, Sanglertsinlapachai D, Techaratsami S, Musikatavorn K, Saoraya J. Machine learning algorithms for early sepsis detection in the emergency department: a retrospective study. Int J Med Inform. 2022 Apr;160:104689.
7. Lanzagorta-Ortega D, Carrillo-Pérez DL, Carrillo-Esper R. Inteligencia artificial en medicina: presente y futuro. Gac Med Mex [Internet]. 2022 [citado el 19 de mayo de 2025];158:17–21. Disponible en: http://www.scielo.org.mx/scielo.php?script=sci_abstract&pid=S0016-38132022001100017&lng=es&nrm=iso&tlng=es
8. Azevedo LCP, Cavalcanti AB, Lisboa T, Pizzol FD, Machado FR. Sepsis is an important healthcare burden in Latin America: a call to action! Rev Bras Ter Intensiva [Internet]. 2018 [citado el 28 de mayo de 2025];30(4):402–4. Disponible en: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6334489/
9. Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O, Peacock R. Storylines of research in diffusion of innovation: a meta-narrative approach to systematic review. Soc Sci Med [Internet]. 2005 Jul 1 [citado el 28 de mayo de 2025];61(2):417–30. Disponible en: https://www.sciencedirect.com/science/article/pii/S0277953604006471
10. Aguayo-Albasini JL, Flores-Pastor B, Soria-Aledo V. Sistema GRADE: clasificación de la calidad de la evidencia y graduación de la fuerza de la recomendación. Cir Esp [Internet]. 2014 Feb 1 [citado el 29 de mayo de 2025];92(2):82–8. Disponible en: http://www.elsevier.es/es-revista-cirugia-espanola-36-articulo-sistemagrade-clasificacion-calidad-evidencia-S0009739X13003394
11. Clinical Questions: PICO and PEO Research | Elsevier Blog [Internet]. [citado el 29 de mayo de 2025]. Disponible en: https://scientific-publishing.webshop.elsevier.com/research-process/clinical-questions-pico-and-peo-research/
12. Wang D, Li J, Sun Y, Ding X, Zhang X, Liu S, et al. A machine learning model for accurate prediction of sepsis in ICU patients. Front Public Health. 2021;9:754348.
13. Ezeobi DP, Musiimenta A, Wasswa W, Kyoyagala S. A neonatal sepsis prediction algorithm using electronic medical record data from Mbarara Regional Referral Hospital. Intell Based Med [Internet]. 2025 Jan 1 [citado el 4 de junio de 2025];11:100198. Disponible en: https://www.sciencedirect.com/science/article/pii/S2666521225000018
14. Li X, Xu X, Xie F, Xu X, Sun Y, Liu X, et al. A time-phased machine learning model for real-time prediction of sepsis in critical care. Crit Care Med. 2020 Oct;48(10):e884–8.
15. Lin TH, Chung HY, Jian MJ, Chang CK, Lin HH, Yen CT, et al. AI-driven innovations for early sepsis detection by combining predictive accuracy with blood count analysis in an emergency setting: retrospective study. J Med Internet Res. 2025 Jan 24;27:e56155.
16. Yang M, Liu C, Wang X, Li Y, Gao H, Liu X, et al. An explainable artificial intelligence predictor for early detection of sepsis. Crit Care Med. 2020 Nov;48(11):e1091–6.
17. Campagner A, Agnello L, Carobene A, Padoan A, Ben FD, Locatelli M, et al. Complete blood count and monocyte distribution width–based machine learning algorithms for sepsis detection: multicentric development and external validation study. J Med Internet Res [Internet]. 2025 Feb 26 [citado el 4 de junio de 2025];27(1):e55492. Disponible en: https://www.jmir.org/2025/1/e55492
18. Ying J, Wang Q, Xu T, Lu Z. Diagnostic potential of a gradient boosting-based model for detecting pediatric sepsis. Genomics [Internet]. 2021 Jan 1 [citado el 4 de junio de 2025];113(1 Pt 2):874–83. Disponible en: https://www.sciencedirect.com/science/article/pii/S0888754320319777
19. Abd El-Aziz RM, Rayan A. Early detection of sepsis using machine learning algorithms. Alex Eng J [Internet]. 2025 Jan [citado el 4 de junio de 2025];111:47–56. Disponible en: https://linkinghub.elsevier.com/retrieve/pii/S1110016824011591
20. Persson I, Macura A, Becedas D, Sjövall F. Early prediction of sepsis in intensive care patients using the machine learning algorithm NAVOY® Sepsis: a prospective randomized clinical validation study. J Crit Care. 2024 Apr;80:154400.
21. Yang AC, Ma WM, Chiang DH, Liao YZ, Lai HY, Lin SC, et al. Early prediction of sepsis using an XGBoost model with single time-point non-invasive vital signs and its correlation with C-reactive protein and procalcitonin: a multi-center study. Intell Based Med [Internet]. 2025 Jan 1 [citado el 4 de junio de 2025];11:100242. Disponible en: https://www.sciencedirect.com/science/article/pii/S2666521225000468
22. He Z, Du L, Zhang P, Zhao R, Chen X, Fang Z. Early sepsis prediction using ensemble learning with deep features and artificial features extracted from clinical electronic health records. Crit Care Med. 2020 Dec;48(12):e1337–42.
23. Yang D, Kim J, Yoo J, Cha WC, Paik H. Identifying the risk of sepsis in patients with cancer using digital health care records: machine learning-based approach. JMIR Med Inform. 2022 Jun 15;10(6):e37689.
24. Hong X, Liu G, Chi Z, Yang T, Zhang Y. Predictive model for urosepsis in patients with upper urinary tract calculi based on ultrasonography and urinalysis using artificial intelligence learning. Int Braz J Urol. 2023;49(2):221–32.
25. Brann F, Sterling NW, Frisch SO, Schrager JD. Sepsis prediction at emergency department triage using natural language processing: retrospective cohort study. JMIR AI [Internet]. 2024 Jan 25 [citado el 4 de junio de 2025];3(1):e49784. Disponible en: https://ai.jmir.org/2024/1/e49784
26. Sakri S, Basheer S, Zain Z, Halimatul N, Nassar D, Alohali M, et al. Sepsis prediction using CNN-BDLSTM and temporal derivatives feature extraction in the IoT medical environment. Comput Mater Contin [Internet]. 2024 [citado el 4 de junio de 2025];79(1):1157–85. Disponible en: https://www.techscience.com/cmc/v79n1/56294
27. Yuan KC, Tsai LW, Lee KH, Cheng YW, Hsu SC, Lo YS, et al. The development of an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit. Int J Med Inform. 2020 Sep;141:104176.
28. Fleuren LM, Klausch TLT, Zwager CL, Schoonmade LJ, Guo T, Roggeveen LF, et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy. Intensive Care Med [Internet]. 2020 [citado el 7 de junio de 2025];46(3):383–400. Disponible en: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7067741/
29. Moor M, Rieck B, Horn M, Jutzeler CR, Borgwardt K. Early prediction of sepsis in the ICU using machine learning: a systematic review. Front Med. 2021;8:607952.
30. Swets JA. Measuring the accuracy of diagnostic systems. Science [Internet]. 1988 Jun 3 [citado el 7 de junio de 2025];240(4857):1285–93. Disponible en: https://www.science.org/doi/10.1126/science.3287615
31. Martínez Pérez JA, Pérez Martín PS. La curva ROC. Med Fam SEMERGEN [Internet]. 2022 Jan 1 [citado el 7 de junio de 2025];49(1). Disponible en: http://www.elsevier.es/es-revista-medicina-familia-semergen-40-articulo-la-curva-roc-S1138359322001952
32. Rodríguez MR, Llopis Roca F, Rubio Díaz R, García DE, Julián Jiménez A. Administración precoz de la antibioterapia empírica en los pacientes atendidos con sospecha de sepsis en el servicio de urgencias: revisión sistemática. Emerg Rev Soc Esp Med Urgenc Emerg [Internet]. 2025 [citado el 7 de junio de 2025];37(1):44–55. Disponible en: https://dialnet.unirioja.es/servlet/articulo?codigo=9915229
33. Silva CC da, Riguetti SL, Vignoli JM, Monnerat MELG, Moura FN de, Glioche RA, et al. Inteligencia artificial en la lucha contra la sepsis: apoyo al diagnóstico y tratamiento clínico. Rev Electron Acervo Saúde. 2025;25(5):e19761. doi:10.25248/reas.e19761.2025
Article Sidebar
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
RHCS uses the Creative Commons License 4.0 International (CC BY 4.0). For more information, please visit