CoChat
Sign Up
Cited by 32Sepsis Diagnosis and TreatmentBacterial Identification and Susceptibility TestingMachine Learning in Healthcare

Machine learning model for the prediction of gram-positive and gram-negative bacterial bloodstream infection based on routine laboratory parameters

Open access
Zhang, Fan, Wang, Hao, Liu, Liyu, Su, Teng, Ji, Bing·2023·BMC Infectious Diseases
DOI 10.1186/s12879-023-08602-4PMC PMC10566101

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Bacterial bloodstream infection is responsible for the majority of cases of sepsis and septic shock. Early recognition of the causative pathogen is pivotal for administration of adequate empiric antibiotic therapy and for the survival of the patients. In this study, we developed a feasible machine learning (ML) model to predict gram-positive and gram-negative bacteremia based on routine laboratory parameters. Methods: Data for 2118 patients with bacteremia were obtained from the Medical Information Mart for Intensive Care dataset. Patients were randomly split into the training set and test set by stratified sampling, and 374 routine laboratory blood test variables were retrieved. Variables with missing values in more than 40% of the patients were excluded. Pearson correlation test was employed to eliminate redundant features. Five ML algorithms were used to build the model based on the selected features. Additionally, 132 patients with bacteremia who were treated at Qilu Hospital of Shandong University were included in an independent test cohort to evaluate the model. Results: After feature selection, 32 variables remained. All the five ML algorithms performed well in terms of discriminating between gram-positive and gram-negative bacteremia, but the performance of convolutional neural network (CNN) and random forest (RF) were better than other three algorithms. Consider of the interpretability of models, RF was chosen for further test (ROC-AUC = 0.768; 95%CI = 0.715–0.798, with a sensitivity of 75.20% and a specificity of 63.79%). To expand the application of the model, a decision tree (DT) was built utilizing the major variables, and it achieved an AUC of 0.679 (95%CI = 0.632–0.723), a sensitivity of 66%, and a specificity of 67.82% in the test cohort. When tested in the Qilu Hospital cohort, the ROC-AUC of the RF and DT models were 0.666 (95%CI = 0.579–0.746) and 0.615 (95%CI = 0.526–0.698), respectively. Finally, a software was developed to make the RF- and DT-based prediction models easily accessible. Conclusion: The present ML-based models could effectively discriminate between gram-positive and gram-negative bacteremia based on routine laboratory blood test results. This simple model would be beneficial in terms of guiding timely antibiotic selection and administration in critically ill patients with bacteremia before their pathogen test results are available.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s12879-023-08602-4

Cites

2 of 2 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Recent studies have explored various ML algorithms, such as logistic regression, random forest, support vector machines, and gradient boosting, to assess the risk of BSIs in real time.

    “Recent studies have explored various ML algorithms, such as logistic regression, random forest, support vector machines, and gradient boosting, to assess the risk of BSIs in real time [10–14].”

    From Early prediction of bloodstream infections in ICU patients using machine learning methods based on routine laboratory parameters (2025) · cites this paper as [12]

    The Background (page unavailable) describes studies using random forest, logistic regression, and support vector machines to predict bloodstream infections or bacteremia. It does not identify gradient boosting among the algorithms, and it does not establish real-time risk assessment. The Discussion (page unavailable) says this paper’s model distinguishes gram-positive from gram-negative bacteremia before pathogen test results are available, rather than assessing whether a patient has a BSI. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • Machine learning models have demonstrated significant potential in predicting BSIs by analyzing readily available clinical and laboratory data.

    “Machine learning models have demonstrated significant potential in predicting BSIs by analyzing readily available clinical and laboratory data.”

    From Early prediction of bloodstream infections in ICU patients using machine learning methods based on routine laboratory parameters (2025) · cites this paper as [12]

  • Most reported ML-based BSI prediction models integrate a combination of vital signs, clinical scores, and laboratory values.

    “However, most reported ML-based BSI prediction models integrate a combination of vital signs, clinical scores, and laboratory values.”

    From Early prediction of bloodstream infections in ICU patients using machine learning methods based on routine laboratory parameters (2025) · cites this paper as [12]

    The Background describes several ML-based bacteremia models but does not identify their inputs as a combination of vital signs, clinical scores, and laboratory values. The Methods > Variables section describes this paper’s own model as using routine laboratory blood tests. Searches for the claimed combination found no passage establishing what most reported models use. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 37

  1. [1]Timsit JF Ruppe E Barbier F Tabah A Bassetti M Bloodstream infections in critically ill patients: an expert statement Intensive Care Med 2020 46 2 266 284 10.1007/s00134-020-05950-6 32047941 PMC7223992 · DOI
  2. [2]Adrie C Garrouste-Orgeas M Ibn Essaied W Schwebel C Darmon M Mourvillier B Attributable mortality of ICU-acquired bloodstream infections: Impact of the source, causative micro-organism, resistance profile and antimicrobial therapy J Infect 2017 74 2 131 141 10.1016/j.jinf.2016.11.001 27838521 · DOI
  3. [3]Zahar JR Timsit JF Garrouste-Orgeas M Francais A Vesin A Descorps-Declere A Outcomes in severe sepsis and patients with septic shock: pathogen species and infection sites are not associated with mortality Crit Care Med 2011 39 8 1886 1895 10.1097/CCM.0b013e31821b827c 21516036 · DOI
  4. [4]Pouwels KB Vansteelandt S Batra R Edgeworth JD Smieszek T Robotham JV Intensive care unit (ICU)-acquired bacteraemia and ICU mortality and discharge: addressing time-varying confounding using appropriate methodology J Hosp Infect 2018 99 1 42 47 10.1016/j.jhin.2017.11.011 29175434 · DOI
  5. [5]Kumar A Ellis P Arabi Y Roberts D Light B Parrillo JE Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock Chest 2009 136 5 1237 1248 10.1378/chest.09-0087 19696123 · DOI
  6. [6]Abe R Oda S Sadahiro T Nakamura M Hirayama Y Tateishi Y Gram-negative bacteremia induces greater magnitude of inflammatory response than Gram-positive bacteremia Crit Care 2010 14 2 R27 10.1186/cc8898 20202204 PMC2887127 · DOI
  7. [7]Bilgili B Haliloglu M Aslan MS Sayan I Kasapoglu US Cinel I Diagnostic accuracy of procalcitonin for differentiating bacteraemic gram-negative sepsis from gram-positive sepsis Turk J Anaesthesiol Reanim 2018 46 1 38 43 30140499 PMC5858887
  8. [8]Liu HH Zhang MW Guo JB Li J Su L Procalcitonin and C-reactive protein in early diagnosis of sepsis caused by either gram-negative or gram-positive bacteria Ir J Med Sci 2017 186 1 207 212 10.1007/s11845-016-1457-z 27139197 · DOI
  9. [9]Xu XJ Tang YM Liao C Song H Yang SL Xu WQ Inflammatory cytokine measurement quickly discriminates gram-negative from gram-positive bacteremia in pediatric hematology/oncology patients with septic shock Intensive Care Med 2013 39 2 319 326 10.1007/s00134-012-2752-4 23179333 · DOI
  10. [10]Debas HT, Donkor P, Gawande A, Jamison DT, Kruk ME, Mock CN, editors. 2015. Essential Surgery. Disease Control Priorities, third edition, volume 1. Washington, DC: World Bank. 10.1596/978-1-4648-0346-8. 26740991 · DOI
  11. [11]Richens JG Lee CM Johri S Improving the accuracy of medical diagnosis with causal machine learning Nat Commun 2020 11 1 3923 10.1038/s41467-020-17419-7 32782264 PMC7419549 · DOI · Paper page
  12. [12]Lynch CJ Liston C New machine-learning technologies for computer-aided diagnosis Nat Med 2018 24 9 1304 1305 10.1038/s41591-018-0178-4 30177823 · DOI
  13. [13]Beeler C Dbeibo L Kelley K Thatcher L Webb D Bah A Assessing patient risk of central line-associated bacteremia via machine learning Am J Infect Control 2018 46 9 986 991 10.1016/j.ajic.2018.02.021 29661634 · DOI
  14. [14]Mahmoud E Al Dhoayan M Bosaeed M Al Johani S Arabi YM Developing machine-learning prediction algorithm for bacteremia in admitted patients Infection and drug resistance 2021 14 757 765 10.2147/IDR.S293496 33658812 PMC7920583 · DOI
  15. [15]Tsai CM Lin CR Zhang H Chiu IM Cheng CY Yu HR Using machine learning to predict bacteremia in febrile children presented to the Emergency Department Diagnostics (Basel, Switzerland) 2020 10 5 307 32429293 PMC7277905
  16. [16]Bhavani SV Lonjers Z Carey KA Afshar M Gilbert ER Shah NS The development and validation of a machine learning model to predict bacteremia and fungemia in hospitalized patients using electronic health record data Crit Care Med 2020 48 11 e1020 e1028 10.1097/CCM.0000000000004556 32796184 PMC7554143 · DOI
  17. [17]Johnson AEW Pollard TJ Shen L Lehman L-wH Feng M Ghassemi M MIMIC-III, a freely accessible critical care database Scientific Data 2016 3 1 160035 10.1038/sdata.2016.35 27219127 PMC4878278 · DOI · Paper page
  18. [19]Goldberger AL Amaral LA Glass L Hausdorff JM Ivanov PC Mark RG PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals Circulation 2000 101 23 E215 E220 10.1161/01.CIR.101.23.e215 10851218 · DOI · Paper page
  19. [20]Hall KK Lyman JA Updated review of blood culture contamination Clin Microbiol Rev 2006 19 4 788 802 10.1128/CMR.00062-05 17041144 PMC1592696 · DOI
  20. [21]Weinstein MP Towns ML Quartey SM Mirrett S Reimer LG Parmigiani G The clinical significance of positive blood cultures in the 1990s: a prospective comprehensive evaluation of the microbiology, epidemiology, and outcome of bacteremia and fungemia in adults Clin Infect Dis 1997 24 4 584 602 10.1093/clind/24.4.584 9145732 · DOI
  21. [22]Muehlematter UJ Mannil M Becker AS Vokinger KN Finkenstaedt T Osterhoff G Vertebral body insufficiency fractures: detection of vertebrae at risk on standard CT images using texture analysis and machine learning Eur Radiol 2019 29 5 2207 2217 10.1007/s00330-018-5846-8 30519934 · DOI
  22. [23]Lundberg SM, Erion, Gabriel G, Lee SI. Consistent individualized feature attribution for tree ensembles. eprint arXiv:180203888. 2018. 10.48550/arXiv.1802.03888. · DOI · Paper page
  23. [24]Cornaglia G Fighting infections due to multidrug-resistant gram-positive pathogens Clin Microbiol Infect 2009 15 3 209 211 10.1111/j.1469-0691.2009.02737.x 19335367 · DOI
  24. [25]Asokan GV Ramadhan T Ahmed E Sanad H WHO global priority pathogens list: a bibliometric analysis of Medline-Pubmed for knowledge mobilization to infection prevention and control practices in Bahrain Oman Med J 2019 34 3 184 193 10.5001/omj.2019.37 31110624 PMC6505350 · DOI · Paper page
  25. [26]Cerceo E Deitelzweig SB Sherman BM Amin AN Multidrug-resistant gram-negative bacterial infections in the hospital setting: overview, implications for clinical practice, and emerging treatment options Microb Drug Resist 2016 22 5 412 431 10.1089/mdr.2015.0220 26866778 · DOI
  26. [27]Garnacho-Montero J Garcia-Garmendia JL Barrero-Almodovar A Jimenez-Jimenez FJ Perez-Paredes C Ortiz-Leyba C Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis Crit Care Med 2003 31 12 2742 2751 10.1097/01.CCM.0000098031.24329.10 14668610 · DOI
  27. [28]Chase M Klasco RS Joyce NR Donnino MW Wolfe RE Shapiro NI Predictors of bacteremia in emergency department patients with suspected infection Am J Emerg Med 2012 30 9 1691 1697 10.1016/j.ajem.2012.01.018 22626814 · DOI
  28. [29]Brodská H Malíčková K Adámková V Benáková H Šťastná MM Zima T Significantly higher procalcitonin levels could differentiate gram-negative sepsis from gram-positive and fungal sepsis Clin Exp Med 2013 13 3 165 170 10.1007/s10238-012-0191-8 22644264 · DOI
  29. [30]Ratzinger F Dedeyan M Rammerstorfer M Perkmann T Burgmann H Makristathis A Neither single nor a combination of routine laboratory parameters can discriminate between gram-positive and gram-negative bacteremia Sci Rep 2015 5 16008 10.1038/srep16008 26522966 PMC4629184 · DOI
  30. [31]Vandijck DM Hoste EA Blot SI Depuydt PO Peleman RA Decruyenaere JM Dynamics of C-reactive protein and white blood cell count in critically ill patients with nosocomial gram positive vs. gram negative bacteremia: a historical cohort study BMC Infect Dis 2007 7 106 10.1186/1471-2334-7-106 17868441 PMC2040151 · DOI
  31. [32]Hoerr V Zbytnuik L Leger C Tam PP Kubes P Vogel HJ Gram-negative and Gram-positive bacterial infections give rise to a different metabolic response in a mouse model J Proteome Res 2012 11 6 3231 3245 10.1021/pr201274r 22483232 PMC3368387 · DOI
  32. [33]Karlsson H Larsson P Wold AE Rudin A Pattern of cytokine responses to gram-positive and gram-negative commensal bacteria is profoundly changed when monocytes differentiate into dendritic cells Infect Immun 2004 72 5 2671 2678 10.1128/IAI.72.5.2671-2678.2004 15102775 PMC387913 · DOI
  33. [34]Sarmiento EU Espiritu BR Gleich GJ Thomas LL IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor potentiate basophil mediator release stimulated by eosinophil granule major basic protein J Immunol 1995 155 4 2211 2221 10.4049/jimmunol.155.4.2211 7543541 · DOI
  34. [35]Michaeli B Martinez A Revelly JP Cayeux MC Chiolero RL Tappy L Effects of endotoxin on lactate metabolism in humans Crit Care 2012 16 4 R139 10.1186/cc11444 22839504 PMC3580724 · DOI
  35. [36]Khatib-Massalha E Bhattacharya S Massalha H Biram A Golan K Kollet O Lactate released by inflammatory bone marrow neutrophils induces their mobilization via endothelial GPR81 signaling Nat Commun 2020 11 1 3547 10.1038/s41467-020-17402-2 32669546 PMC7363928 · DOI
  36. [37]Tapia P Soto D Bruhn A Alegria L Jarufe N Luengo C Impairment of exogenous lactate clearance in experimental hyperdynamic septic shock is not related to total liver hypoperfusion Crit Care 2015 19 188 10.1186/s13054-015-0928-3 25898244 PMC4432956 · DOI
  37. [38]Jirillo E Caccavo D Magrone T Piccigallo E Amati L Lembo A The role of the liver in the response to LPS: experimental and clinical findings J Endotoxin Res 2002 8 5 319 327 12537690

Cited by

Sorted by citations

Authors
Zhang, F., Wang, H., Liu, L., Su, T., & Ji, B.
Affiliations
Shandong University (CN)
Corresponding author
Ji, Bing
ORCID
Venue
BMC Infectious Diseases
Publisher
Springer Science and Business Media LLC
Published
2023-10-10
Volume
23
Issue
1
Pages
675-675
ISSN
1471-2334 (ISSN-L)
DOI
10.1186/s12879-023-08602-4
PMC
PMC10566101
OpenAlex
W4387473747
Cited by
32
References
2
Citations by year
2026: 10 · 2025: 12 · 2024: 10
FWCI
6.39
Type
journal article
Language
en
Open access
Licence
Locations
  • BMC Infectious Diseases (publishedVersion, cc-by)
  • Research Square (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Natural Science Foundation of China (81873927, 62173212)
  • Taishan Scholars Program of Shandong Province (tsqn202103165)
  • Clinical Research Center of Shandong University (2020SDUCRCC013)
Topics
  • Health Sciences › Medicine › Epidemiology › Sepsis Diagnosis and Treatment
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Clinical Biochemistry › Bacterial Identification and Susceptibility Testing
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
Keywords
gram-positive bacteremia, gram-negative bacteremia, bacterial bloodstream infection, pathogen identification, routine laboratory parameters, random forest, decision tree, convolutional neural networks, sepsis, septic shock, empirical antibiotic therapy, MIMIC-III, forecasting software, feature selection, interpretability, sepsis severity score, APACHE II score, SOFA score
MeSH terms
Anti-Bacterial Agents, Humans, Shock, Septic, Bacteremia, Gram-Negative Bacterial Infections, Sepsis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review