An explainable machine learning model for early pediatric sepsis prediction using cytokine and routine laboratory data

Machine Learning


  • Matot, I. & Sprung, C. L. Definition of sepsis. Intensive Care Med. 27, S3–S9 (2001).

    Article 
    PubMed 

    Google Scholar 

  • Watson, R. S. et al. The burden and contemporary epidemiology of sepsis in children. Lancet Child Adolesc. Health 8, 670–681 (2024).

    Article 
    PubMed 

    Google Scholar 

  • Rudd, K. E. et al. Global, regional, and national sepsis incidence and mortality, 1990-2017: analysis for the Global Burden of Disease Study. Lancet 395, 200–211 (2020).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Plunkett, A. & Tong, J. Sepsis in children. BMJ 350, h3017 (2015).

    Article 
    PubMed 

    Google Scholar 

  • Weiss, S. L. et al. Delayed antimicrobial therapy increases mortality and organ dysfunction duration in pediatric sepsis. Crit. Care Med. 42, 2409–2417 (2014).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Póvoa, P. et al. How to use biomarkers of infection or sepsis at the bedside: guide to clinicians. Intensive Care Med. 49, 142–153 (2023).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Pierrakos, C. & Vincent, J. L. Sepsis biomarkers: a review. Crit. Care 14, R15 (2010).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Faix, J. D. Biomarkers of sepsis. Crit. Rev. Clin. Lab. Sci. 50, 23–36 (2013).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Barichello, T., Generoso, J. S., Singer, M. & Dal-Pizzol, F. Biomarkers for sepsis: more than just fever and leukocytosis—a narrative review. Crit. Care 26, 14 (2022).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Pierrakos, C., Velissaris, D., Bisdorff, M., Marshall, J. C. & Vincent, J. L. Biomarkers of sepsis: time for a reappraisal. Crit. Care 24, 287 (2020).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Alpern, E. R. et al. Derivation and validation of predictive models for early pediatric sepsis. JAMA Pediatr. 179, 1318–1325 (2025).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Chanci, D. et al. Machine learning model for daily prediction of pediatric sepsis using Phoenix criteria. Pediatr. Res. 99, 631–637 (2026).

    Article 
    PubMed 

    Google Scholar 

  • Desautels, T. et al. Prediction of sepsis in the intensive care unit with minimal electronic health record data: a machine learning approach. JMIR Med. Inform. 4, e28 (2016).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Flannery, A. H. et al. Frequency and types of healthcare encounters in the week preceding a sepsis hospitalization: a systematic review. Crit. Care Explor. 4, e0635 (2022).

    PubMed 
    PubMed Central 

    Google Scholar 

  • Loots, F. J. et al. Management of sepsis in out-of-hours primary care: a retrospective study of patients admitted to the intensive care unit. BMJ Open 8, e022832 (2018).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Gerard, R. et al. Is “ pre-sepsis “ the new sepsis? A narrative review. PLoS Pathog. 21, e1013372 (2025).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Seymour, C. W. et al. Time to treatment and mortality during mandated emergency care for sepsis. N. Engl. J. Med. 376, 2235–2244 (2017).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Goldstein, B., Giroir, B. & Randolph, A.International Consensus Conference on Pediatric Sepsis International Pediatric Sepsis Consensus Conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr. Crit. Care Med. 6, 2–8 (2005).

    Article 
    PubMed 

    Google Scholar 

  • Schlapbach, L. J. et al. International consensus criteria for pediatric sepsis and septic shock. JAMA 331, 665–674 (2024).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Zhang, Q. et al. Circulating mitochondrial DAMPs cause inflammatory responses to injury. Nature 464, 104–107 (2010).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • van der Poll, T., van de Veerdonk, F. L., Scicluna, B. P. & Netea, M. G. The immunopathology of sepsis and potential therapeutic targets. Nat. Rev. Immunol. 17, 407–420 (2017).

    Article 
    PubMed 

    Google Scholar 

  • Wynn, T. A. Type 2 cytokines: mechanisms and therapeutic strategies. Nat. Rev. Immunol. 15, 271–282 (2015).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Chousterman, B. G., Swirski, F. K. & Weber, G. F. Cytokine storm and sepsis disease pathogenesis. Semin. Immunopathol. 39, 517–528 (2017).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Zhang, Y., Li, B. & Ning, B. Evaluating IL-6 and IL-10 as rapid diagnostic tools for gram-negative bacteria and as disease severity predictors in pediatric sepsis patients in the intensive care unit. Front. Immunol. 13, 1043968 (2022).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Zhou, X. et al. Combining host immune response biomarkers and clinical scores for early prediction of sepsis in infection patients. Ann. Med. 56, 2396569 (2024).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Freitas, A. et al. IL-17 receptor signaling is required to control polymicrobial sepsis. J. Immunol. 182, 7846–7854 (2009).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Tangye, S. G. & Puel, A. The Th17/IL-17 axis and host defense against fungal infections. J. Allergy Clin. Immunol. Pract. 11, 1624–1634 (2023).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Liew, F. Y. IL-33: a Janus cytokine. Ann. Rheum. Dis. 71, i101–i104 (2012).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Alves-Filho, J. C. et al. Interleukin-33 attenuates sepsis by enhancing neutrophil influx to the site of infection. Nat. Med. 16, 708–712 (2010).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Hotchkiss, R. S. et al. Sepsis and septic shock. Nat. Rev. Dis. Primers 2, 16045 (2016).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Payen, D. et al. Multicentric experience with interferon gamma therapy in sepsis induced immunosuppression. A case series. BMC Infect. Dis. 19, 931 (2019).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Kawai, T. & Akira, S. The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat. Immunol. 11, 373–384 (2010).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Takahama, M. et al. A pairwise cytokine code explains the organism-wide response to sepsis. Nat. Immunol. 25, 226–239 (2024).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Palmowski, L. et al. Mortality-associated plasma proteome dynamics in a prospective multicentre sepsis cohort. EBioMedicine 111, 105508 (2025).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Yan, F. et al. Association between the stress hyperglycemia ratio and 28-day all-cause mortality in critically ill patients with sepsis: a retrospective cohort study and predictive model establishment based on machine learning. Cardiovasc. Diabetol. 23, 163 (2024).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Peduzzi, P., Concato, J., Kemper, E., Holford, T. R. & Feinstein, A. R. A simulation study of the number of events per variable in logistic regression analysis. J. Clin. Epidemiol. 49, 1373–1379 (1996).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Collins, G. S. et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ 385, e078378 (2024).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Moons, K. G. M. et al. PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods. BMJ 388, e082505 (2025).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Van Calster, B. et al. Reporting and interpreting decision curve analysis: a guide for investigators. Eur. Urol. 74, 796–804 (2018).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Akhondi-Asl, A., Geva, A., Burns, J. P. & Mehta, N. M. Dynamic prediction of mortality using longitudinally measured pediatric Sequential Organ Failure Assessment scores: a joint modeling approach. Pediatr. Crit. Care Med. 25, 443–451 (2024).

    Article 
    PubMed 

    Google Scholar 

  • Zhang, L. D. et al. pSOFA score serves as predicting the severity and outcome of sepsis in critically ill children. Sci. Rep. 15, 43595 (2025).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Balamuth, F. et al. Validation of the pediatric Sequential Organ Failure Assessment score and evaluation of third international consensus definitions for sepsis and septic shock definitions in the pediatric emergency department. JAMA Pediatr. 176, 672–678 (2022).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Hadzhieva-Hristova, A., Krumova, D., Stoeva, T., Georgieva, R. & Iotova, V. Assessment of Phoenix Sepsis Score, pSOFA, PELOD-2, and PRISM III in pediatric intensive care. Children 12, 262 (2025).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • AbdelRazic, M. I., Hakeem, G. L. A., Hanna, M. S., Mohamed, O. M. & Abuelela, I. S. Enhanced mortality prediction in pediatric sepsis using NGAL: a comparison with PRISM III scores in critical care settings. Eur. J. Pediatr. 184, 201 (2025).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Matics, T. J. & Sanchez-Pinto, L. N. Adaptation and validation of a pediatric Sequential Organ Failure Assessment score and evaluation of the Sepsis-3 definitions in critically ill children. JAMA Pediatr. 171, e172352 (2017).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Tan, X. et al. Constructing a predictive model for early-onset sepsis in neonatal intensive care unit newborns based on SHapley Additive exPlanations explainable machine learning. Transl. Pediatr. 13, 1933–1946 (2024).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Fleuren, L. M. et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy. Intensive Care Med. 46, 383–400 (2020).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Tennant, R. et al. A scoping review on pediatric sepsis prediction technologies in healthcare. npj Digit. Med. 7, 353 (2024).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Shi, J., Hubbard, A. E., Fong, N. & Pirracchio, R. Implicit bias in ICU electronic health record data: measurement frequencies and missing data rates of clinical variables. BMC Med. Inform. Decis. Mak. 25, 241 (2025).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Gottesman, O. et al. Guidelines for reinforcement learning in healthcare. Nat. Med. 25, 16–18 (2019).

    Article 
    CAS 
    PubMed 

    Google Scholar 

  • Amann, J. et al. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med. Inform. Decis. Mak. 20, 310 (2020).

    Article 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Vergadi, E. et al. An IL-10/DEL-1 axis supports granulopoiesis and survival from sepsis in early life. Nat. Commun. 15, 680 (2024).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Xu, H., Turnquist, H. R., Hoffman, R. & Billiar, T. R. Role of the IL-33-ST2 axis in sepsis. Mil. Med. Res. 4, 3 (2017).

    PubMed 
    PubMed Central 

    Google Scholar 

  • Zhou, Y., Xu, Z. & Liu, Z. Role of the IL-33-ST2 pathway in regulating inflammation: current evidence and future perspectives. J. Transl. Med. 21, 902 (2023).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Chen, X., Yuan, S., Mi, L., Long, Y. & He, H. Pannexin1: insight into inflammatory conditions and its potential involvement in multiple organ dysfunction syndrome. Front. Immunol. 14, 1217366 (2023).

    Article 
    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 



  • Source link