Transferability and interpretability of the sepsis prediction models in the intensive care unit
Open accessPublished version of a preprint
The DOI matches its Crossref record; 47 of the 47 references it cites were found in a registry or an index.
Abstract
Background: We aimed to develop an early warning system for real-time sepsis prediction in the ICU by machine learning methods, with tools for interpretative analysis of the predictions. In particular, we focus on the deployment of the system in a target medical center with small historical samples. Methods: Light Gradient Boosting Machine (LightGBM) and multilayer perceptron (MLP) were trained on Medical Information Mart for Intensive Care (MIMIC-III) dataset and then finetuned on the private Historical Database of local Ruijin Hospital (HDRJH) using transfer learning technique. The Shapley Additive Explanations (SHAP) analysis was employed to characterize the feature importance in the prediction inference. Ultimately, the performance of the sepsis prediction system was further evaluated in the real-world study in the ICU of the target Ruijin Hospital. Results: The datasets comprised 6891 patients from MIMIC-III, 453 from HDRJH, and 67 from Ruijin real-world data. The area under the receiver operating characteristic curves (AUCs) for LightGBM and MLP models derived from MIMIC-III were 0.98 − 0.98 and 0.95 − 0.96 respectively on MIMIC-III dataset, and, in comparison, 0.82 − 0.86 and 0.84 − 0.87 respectively on HDRJH, from 1 to 5 h preceding. After transfer learning and ensemble learning, the AUCs of the final ensemble model were enhanced to 0.94 − 0.94 on HDRJH and to 0.86 − 0.9 in the real-world study in the ICU of the target Ruijin Hospital. In addition, the SHAP analysis illustrated the importance of age, antibiotics, net balance, and ventilation for sepsis prediction, making the model interpretable. Conclusions: Our machine learning model allows accurate real-time prediction of sepsis within 5-h preceding. Transfer learning can effectively improve the feasibility to deploy the prediction model in the target cohort, and ameliorate the model performance for external validation. SHAP analysis indicates that the role of antibiotic usage and fluid management needs further investigation. We argue that our system and methodology have the potential to improve ICU management by helping medical practitioners identify at-sepsis-risk patients and prepare for timely diagnosis and intervention. Trial registration : NCT05088850 (retrospectively registered).
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures



Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Furthermore, multi-centre validation and prospective testing remain rare [12, 13, 18], making it difficult to identify which interpretability techniques are most reliable in real-world practice.”
From Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“These interpretability strategies enabled the identification of key physiological variables driving model predictions, such as respiratory rate (RR), Glasgow Coma Scale, mean arterial pressure (MAP), lactate, creatinine, heart rate (HR), blood urea nitrogen (BUN), body temperature, platelet count, and urine output [7, 12–18, 22, 24, 25, 39].”
From Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Only a limited number of studies conducted external or temporal validation ([7, 15, 18, 22, 24]), and very few models have been prospectively implemented in real clinical environments.”
From Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Most studies relied on retrospective datasets, including MIMIC-III, MIMIC-IV, and other ICU databases, with cohort sizes ranging from 4,500 to 14,500 patients ([12, 13, 18, 19, 22, 23]).”
From Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction (2026) · cites this paper as [18]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Interpretability methods were widely applied across these studies, with SHAP emerging as the most frequently used technique [12, 13, 15, 16, 18, 19, 22, 24, 39].”
From Explainable AI for critical care: a systematic review of interpretable models for sepsis and ICU mortality prediction (2026) · cites this paper as [18]
Experimental — accuracy not yet measured
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569
The third international consensus definitions for sepsis and septic shock (Sepsis-3).
Singer, M., Deutschman, C. S., Seymour, C. W., Shankar-Hari, M., Annane, D., Bauer, M., Bellomo, R., Bernard, G. R., Chiche, J.-D., Coopersmith, C. M., et al. (2016) · JAMA · doi:10.1001/jama.2016.0287 · Cited by 28,250
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707
LightGBM: A Highly Efficient Gradient Boosting Decision Tree
(2017) · HAL · Cited by 10,512
Angus, D. C., Linde-Zwirble, W. T., Lidicker, J., Clermont, G., Carcillo, J., & Pinsky, M. R. (2001) · Critical Care Medicine · doi:10.1097/00003246-200107000-00002 · Cited by 8,503
MIMIC-III, a freely accessible critical care database
Johnson, A. E. W., Pollard, T. J., Shen, L., Lehman, L.-W. H., Feng, M., Ghassemi, M., Moody, B., Szolovits, P., Anthony Celi, L., & Mark, R. G. (2016) · Scientific Data · doi:10.1038/sdata.2016.35 · Cited by 8,141
Proceedings of the 21st International Conference on Neural Information Processing Systems
Koller, D., Schuurmans, D., Bengio, Y., & Bottou, L. (2008) · Cited by 4,461
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. M. (2015) · BMJ · doi:10.1136/bmj.g7594 · Cited by 4,388 · cited 1×
“Our study complies with the relevant reporting guidelines, namely the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement [24].”
MethodCollins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. M. (2015) · Annals of Internal Medicine · doi:10.7326/m14-0697 · Cited by 3,267
Cecconi, M., Evans, L., Levy, M., & Rhodes, A. (2018) · The Lancet · doi:10.1016/s0140-6736(18)30696-2 · Cited by 2,254
Pollard, T. J., Johnson, A. E. W., Raffa, J. D., Celi, L. A., Mark, R. G., & Badawi, O. (2018) · Scientific Data · doi:10.1038/sdata.2018.178 · Cited by 1,993
The effect of age on the development and outcome of adult sepsis*
Martin, G. S., Mannino, D. M., & Moss, M. (2006) · Critical Care Medicine · doi:10.1097/01.ccm.0000194535.82812.ba · Cited by 1,040
An interpretable machine learning model for accurate prediction of sepsis in the ICU
Nemati, S., Holder, A., Razmi, F., Stanley, M. D., Clifford, G. D., & Buchman, T. G. (2018) · Critical Care Medicine · doi:10.1097/ccm.0000000000002936 · Cited by 832 · cited 1×
“Machine learning has emerged as a promising tool for the early detection of sepsis occurrence based on electronic medical records, laboratory data, and biomedical signals (6-14).”
MethodA targeted real-time early warning score (TREWScore) for septic shock.
Henry, K. E., Hager, D. N., Pronovost, P. J., & Saria, S. (2015) · Science Translational Medicine · doi:10.1126/scitranslmed.aab3719 · Cited by 598 · cited 2×
“Machine learning has been considered a promising method for sepsis prediction in the ICU [6–16 , 20– 22 ].”
MethodDesautels, T., Calvert, J., Hoffman, J., Jay, M., Kerem, Y., Shieh, L., Shimabukuro, D., Chettipally, U., Feldman, M. D., Barton, C., et al. (2016) · JMIR Medical Informatics · doi:10.2196/medinform.5909 · Cited by 509 · cited 1×
“Machine learning has emerged as a promising tool for the early detection of sepsis occurrence based on electronic medical records, laboratory data, and biomedical signals (6-14).”
MethodA scoping review of transfer learning research on medical image analysis using ImageNet
Morid, M. A., Borjali, A., & Del Fiol, G. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2020.104115 · Cited by 508
Siontis, G. C. M., Tzoulaki, I., Castaldi, P. J., & Ioannidis, J. P. A. (2015) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2014.09.007 · Cited by 459
Antibiotic stewardship in the intensive care unit
Luyt, C.-E., Bréchot, N., Trouillet, J.-L., & Chastre, J. (2014) · Critical Care · doi:10.1186/s13054-014-0480-6 · Cited by 388
Shimabukuro, D. W., Barton, C. W., Feldman, M. D., Mataraso, S. J., & Das, R. (2017) · BMJ Open Respiratory Research · doi:10.1136/bmjresp-2017-000234 · Cited by 382
Mao, Q., Jay, M., Hoffman, J. L., Calvert, J., Barton, C., Shimabukuro, D., Shieh, L., Chettipally, U., Fletcher, G., Kerem, Y., et al. (2018) · BMJ Open · doi:10.1136/bmjopen-2017-017833 · Cited by 336
Klein Klouwenberg, P. M. C., Cremer, O. L., van Vught, L. A., Ong, D. S. Y., Frencken, J. F., Schultz, M. J., Bonten, M. J., & van der Poll, T. (2015) · Critical Care · doi:10.1186/s13054-015-1035-1 · Cited by 281
Rationalizing antimicrobial therapy in the ICU: a narrative review
Timsit, J.-F., Bassetti, M., Cremer, O., Daikos, G., de Waele, J., Kallil, A., Kipnis, E., Kollef, M., Laupland, K., Paiva, J.-A., et al. (2019) · Intensive Care Medicine · doi:10.1007/s00134-019-05520-5 · Cited by 266
Ayres, J. S., Trinidad, N. J., & Vance, R. E. (2012) · Nature Medicine · doi:10.1038/nm.2729 · Cited by 213
Early detection of sepsis utilizing deep learning on electronic health record event sequences
Lauritsen, S. M., Kalør, M. E., Kongsgaard, E. L., Lauritsen, K. M., Jørgensen, M. J., Lange, J., & Thiesson, B. (2020) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2020.101820 · Cited by 190
Barton, C., Chettipally, U., Zhou, Y., Jiang, Z., Lynn-Palevsky, A., Le, S., Calvert, J., & Das, R. (2019) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2019.04.027 · Cited by 175
Filbin, M. R., Lynch, J., Gillingham, T. D., Thorsen, J. E., Pasakarnis, C. L., Nepal, S., Matsushima, M., Rhee, C., Heldt, T., & Reisner, A. T. (2018) · Critical Care Medicine · doi:10.1097/ccm.0000000000003260 · Cited by 167
McCoy, A., & Das, R. (2017) · BMJ Open Quality · doi:10.1136/bmjoq-2017-000158 · Cited by 148
Hanke, M., Adelhöfer, N., Kottke, D., Iacovella, V., Sengupta, A., Kaule, F. R., Nigbur, R., Waite, A. Q., Baumgartner, F., & Stadler, J. (2016) · Scientific Data · doi:10.1038/sdata.2016.92 · Cited by 127
Brotfain, E., Koyfman, L., Toledano, R., Borer, A., Fucs, L., Galante, O., Frenkel, A., Kutz, R., & Klein, M. (2016) · The American Journal of Emergency Medicine · doi:10.1016/j.ajem.2016.07.058 · Cited by 114
Wardi, G., Carlile, M., Holder, A., Shashikumar, S., Hayden, S. R., & Nemati, S. (2021) · Annals of Emergency Medicine · doi:10.1016/j.annemergmed.2020.11.007 · Cited by 96
Kenzaka, T., Okayama, M., Kuroki, S., Fukui, M., Yahata, S., Hayashi, H., Kitao, A., Sugiyama, D., Kajii, E., & Hashimoto, M. (2012) · Internal Medicine · doi:10.2169/internalmedicine.51.6951 · Cited by 90
Detecting Patient Deterioration Using Artificial Intelligence in a Rapid Response System
Cho, K.-J., Kwon, O., Kwon, J.-M., Lee, Y., Park, H., Jeon, K.-H., Kim, K.-H., Park, J., & Oh, B.-H. (2020) · Critical Care Medicine · doi:10.1097/ccm.0000000000004236 · Cited by 88
MGP-AttTCN: An interpretable machine learning model for the prediction of sepsis
Rosnati, M., & Fortuin, V. (2021) · PLoS ONE · doi:10.1371/journal.pone.0251248 · Cited by 81
Gandin, I., Scagnetto, A., Romani, S., & Barbati, G. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103876 · Cited by 81
Burdick, H., Pino, E., Gabel-Comeau, D., McCoy, A., Gu, C., Roberts, J., Le, S., Slote, J., Pellegrini, E., Green-Saxena, A., et al. (2020) · BMJ Health & Care Informatics · doi:10.1136/bmjhci-2019-100109 · Cited by 80
Matsubara, T., Yamakawa, K., Umemura, Y., Gando, S., Ogura, H., Shiraishi, A., Kushimoto, S., Abe, T., Tarui, T., Hagiwara, A., et al. (2019) · Thrombosis Research · doi:10.1016/j.thromres.2019.07.002 · Cited by 79
A deep learning approach for sepsis monitoring via severity score estimation
Aşuroğlu, T., & Oğul, H. (2020) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2020.105816 · Cited by 78
An interpretable deep-learning model for early prediction of sepsis in the emergency department
Zhang, D., Yin, C., Hunold, K. M., Jiang, X., Caterino, J. M., & Zhang, P. (2021) · Patterns · doi:10.1016/j.patter.2020.100196 · Cited by 76
Persson, I., Östling, A., Arlbrandt, M., Söderberg, J., & Becedas, D. (2021) · JMIR Formative Research · doi:10.2196/28000 · Cited by 58 · cited 1×
“Machine learning has emerged as a promising tool for the early detection of sepsis occurrence based on electronic medical records, laboratory data, and biomedical signals [6–14].”
MethodLiu, K., Zhang, X., Chen, W., Yu, A. S. L., Kellum, J. A., Matheny, M. E., Simpson, S. Q., Hu, Y. Q., & Liu, M. (2022) · JAMA Network Open · doi:10.1001/jamanetworkopen.2022.19776 · Cited by 56
Chen, H., Lundberg, S. M., Erion, G., Kim, J. H., & Lee, S.-I. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00536-y · Cited by 48
Marik, P. E., & Farkas, J. D. (2018) · Critical Care Medicine · doi:10.1097/ccm.0000000000003310 · Cited by 43
Customization of a Severity of Illness Score Using Local Electronic Medical Record Data
Lee, J., & Maslove, D. M. (2017) · Journal of Intensive Care Medicine · doi:10.1177/0885066615585951 · Cited by 27
Cross-Domain Transfer Learning for PCG Diagnosis Algorithm
Tseng, K.-K., Wang, C., Huang, Y.-F., et al. (2021) · Biosensors · doi:10.3390/bios11040127 · Cited by 26
Self-citationSEPRES: Intensive Care Unit Clinical Data Integration System to Predict Sepsis
Chen, Q., Li, R., Lin, C., Lai, C., Huang, Y., Lu, W., & Li, L. (2022) · Applied Clinical Informatics · doi:10.1055/a-1990-3037 · Cited by 9
Self-citationSEPRES: Sepsis prediction via the clinical data integration system in the ICUPreprint
Chen, Q., Li, R., Lin, C., Lai, C., Huang, Y., Lu, W., & Li, L. (2022) · medRxiv · doi:10.1101/2022.07.06.22277188 · Cited by 3
Self-citationLin, T., Dhillon, N. K., Conde, G., Toscano, S., Margulies, D. R., Barmparas, G., & Ley, E. J. (2020) · The American Journal of Surgery · doi:10.1016/j.amjsurg.2020.08.032 · Cited by 2
Research works printed but not linked: 1
- Moor M, Horn M, Rieck B, Roqueiro D, Borgwardt K. Early recognition of sepsis with Gaussian process temporal convolutional networks and dynamic time warping. Mach Learn Healthc Conf PMLR. 2019;106:2–26.
Cited by
AI Advances in ICU with an Emphasis on Sepsis Prediction: An Overview
Stylianides, C., Nicolaou, A., Sulaiman, W. A., Alexandropoulou, C.-A., Panagiotopoulos, I., Karathanasopoulou, K., Dimitrakopoulos, G., Kleanthous, S., Politi, E., Ntalaperas, D., et al. (2025) · Machine Learning and Knowledge Extraction · doi:10.3390/make7010006 · Cited by 28
“For sepsis prediction, most studies use ML built-in feature importance [112,117,119,124,129,135,136,141,232,233] and SHAP [119,128,233–236].”
Athukorala, V. S., & Ilmini, W. M. K. S. (2026) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-026-03344-0 · Cited by 10
“For instance, Chen et al. [18] evaluated sepsis prediction models across multiple hospitals and found that interpretable models consistently outperformed opaque systems when applied in new clinical settings.”
Influential
Metadata
- Authors
- Chen, Q., Li, R., Lin, C., Lai, C., Chen, D., Qu, H., Huang, Y., Lu, W., Tang, Y., & Li, L.
- Affiliations
- Fudan University (CN)
- Shanghai Jiao Tong University (CN)
- Shanghai Electric (China) (CN)
- Corresponding author
- Li, Lei
- Venue
- BMC Medical Informatics and Decision Making
- Published
- 2022-12-29
- Volume
- 22
- Issue
- 1
- Pages
- 343-343
- ISSN
- 1472-6947 (ISSN-L)
- PMC
- PMC9798724
- OpenAlex
- W4313242224
- Cited by
- 28
- References
- 47
- Citations by year
- 2026: 9 · 2025: 7 · 2024: 9 · 2023: 1 · 2022: 2
- FWCI
- 3.43
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- BMC Medical Informatics and Decision Making (publishedVersion, cc-by)
- medRxiv (acceptedVersion)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Funders
- the Science and Technology Commission of Shanghai Municipality (19JC1420101)
- Shanghai Municipal Science and Technology Major Project, the ZHANGJIANG LAB (2018SHZDZX01)
- Topics
- Health Sciences › Medicine › Critical Care and Intensive Care Medicine › Sepsis Diagnosis and Treatment
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Keywords
- sepsis prediction, intensive care unit, early warning systems, transfer learning, model interpretability, SHAP, LightGBM, multilayer perceptron, ensemble learning, MIMIC-III, antibiotic use, fluid management, real-time prediction, external validation, net balance, mechanical ventilation, age as risk factor
- MeSH terms
- Critical Care, Humans, Intensive Care Units, Databases, Factual, Sepsis, Area Under Curve
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.