AI Advances in ICU with an Emphasis on Sepsis Prediction: An Overview
Open accessThe DOI matches its Crossref record; 243 of the 243 references it cites were found in a registry or an index.
Abstract
Artificial intelligence (AI) is increasingly applied in a wide range of healthcare and Intensive Care Unit (ICU) areas to serve—among others—as a tool for disease detection and prediction, as well as for healthcare resources’ management. Since sepsis is a high mortality and rapidly developing organ dysfunction disease afflicting millions in ICUs and costing huge amounts to treat, the area can benefit from the use of AI tools for early and informed diagnosis and antibiotic administration. Additionally, resource allocation plays a crucial role when patient flow is increased, and resources are limited. At the same time, sensitive data use raises the need for ethical guidelines and reflective datasets. Additionally, explainable AI is applied to handle AI opaqueness. This study aims to present existing clinical approaches for infection assessment in terms of scoring systems and diagnostic biomarkers, along with their limitations, and an extensive overview of AI applications in healthcare and ICUs in terms of (a) sepsis detection/prediction and sepsis mortality prediction, (b) length of ICU/hospital stay prediction, and (c) ICU admission/hospitalization prediction after Emergency Department admission, each constituting an important factor towards either prompt interventions and improved patient wellbeing or efficient resource management. Challenges of AI applications in ICU are addressed, along with useful recommendations to mitigate them. Explainable AI applications in ICU are described, and their value in validating, and translating predictions in the clinical setting is highlighted. The most important findings and future directions including multimodal data use and Transformer-based models are discussed. The goal is to make research in AI advances in ICU and particularly sepsis prediction more accessible and provide useful directions on future work.
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
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.
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
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 · cited 1×
“It is also used in the definition of sepsis [97], while its ‘derivative’, namely Quick-SOFA, is used as a screening tool to identify sepsis in patients [84].”
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Barredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423 · cited 1×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
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 · cited 2×
“Data are getting larger as interoperability increases, more countries and clinical sites share their data [159,188], and dataset(s) are being updated [147,187].”
Le Gall, J.-R. (1993) · JAMA: The Journal of the American Medical Association · doi:10.1001/jama.270.24.2957 · Cited by 7,773 · cited 1×
“SAPS II was validated only in North America and Europe and excluded patients below or at the age of 16 years old, as well as those with burn injuries, coronary artery disease, or a history of cardiac surgery [84,85,88,94,101–103].”
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz682 · Cited by 7,218 · cited 1×
“To improve model performance, future work is expected to investigate sepsis prediction and other clinical outcomes using multimodal data, Transformer-based models, specific disease cohorts, and be informed and driven by clinical knowledge.”
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01228-7 · Cited by 6,909 · cited 1×
“Model-specific methods (e.g., Gradient-weighted Class Activation Mapping-Grad-CAM [229], Sensitivity Analysis [230], Heatmaps [221,227], Instance-wise variable selection (INVASE) [231], and belief rule-based inference methodology [228]) are applied to ML or DL models with a specific structure or…”
Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework
(2001) · Clinical Pharmacology & Therapeutics · doi:10.1067/mcp.2001.113989 · Cited by 6,762
MIMIC-IV, a freely accessible electronic health record dataset
Johnson, A. E. W., Bulgarelli, L., Shen, L., Gayles, A., Shammout, A., Horng, S., Pollard, T. J., Hao, S., Moody, B., Gow, B., et al. (2023) · Scientific Data · doi:10.1038/s41597-022-01899-x · Cited by 3,401 · cited 2×
“Data are getting larger as interoperability increases, more countries and clinical sites share their data [159,188], and dataset(s) are being updated [147,187].”
Towards A Rigorous Science of Interpretable Machine Learning
Doshi-Velez, F., & Kim, B. (2017) · arXiv · doi:10.48550/arxiv.1702.08608 · Cited by 3,121 · cited 2×
“Explainability or the area of Explainable Artificial Intelligence (XAI) [3,4] aims to achieve confidence, trustworthiness, accessibility, causality, and transferability in predictions, so that health professionals can understand and correlate the results with the clinical practice [5,6].”
Research works printed but not linked: 12
- Isravel, 2020, Improved heart disease diagnostic IoT model using machine learning techniques, Neuroscience, 9, 4442
- Gupta, 2021, AI-Enabled COVID-19 Outbreak Analysis and Prediction: Indian States Vs. Union Territories, Comput. Mater. Contin., 67, 933
- Flores, 2001, C reactive protein as marker of infection among patients with severe closed trauma, Enfermedades Infecc. Microbiol. Clin., 19, 61
- Ayazoglu, 2011, A comparison of APACHE II and APACHE IV scoring systems in predicting outcome in patients admitted with stroke to an intensive care unit, Anaesth. Pain Intensiv. Care, 15, 7
- Yang, J., Karstens, L., Ross, C., and Yala, A. (2022). AI gone astray: Technical supplement. arXiv.
- Rocheteau, E., Tong, C., Veličković, P., Lane, N., and Liò, P. (2021). Predicting patient outcomes with graph representation learning. arXiv.
- Wen, 2022, Time-to-event modeling for hospital length of stay prediction for COVID-19 patients, Mach. Learn. Appl., 9, 100365
- Ahmed, 2024, Developing a decision model to early predict ICU admission for COVID-19 patients: A machine learning approach, Intell. Med., 9, 100136
- Lundberg, S., and Lee, S.-I. (2017). SHAP: A Unified Approach to Interpreting Model Predictions. Adv. Neural Inf. Process. Syst., 1–10.
- Pick, F. (2021). Explainable Machine Learning for Predicting Sepsis Outcome. [Master’s Thesis, Department Computer Science at Swansea University].
- Singhal, K., Tu, T., Gottweis, J., Sayres, R., Wulczyn, E., Hou, L., Clark, K., Pfohl, S., Cole-Lewis, H., and Neal, D. (2023). Towards expert-level medical question answering with large language models. arXiv.
- Team, G., Georgiev, P., Lei, V.I., Burnell, R., Bai, L., Gulati, A., Tanzer, G., Vincent, D., Pan, Z., and Wang, S. (2024). Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv.
Cited by
Abbas, Q., Jeong, W., & Lee, S. W. (2025) · Healthcare · doi:10.3390/healthcare13172154 · Cited by 236
“This not only aids in clinical interpretation but also aligns AI recommendations with clinical reasoning, thereby increasing user trust and adoption [24,25].”
Transforming sepsis management: AI-driven innovations in early detection and tailored therapies
Papareddy, P., Lobo, T. J., Holub, M., Bouma, H., Maca, J., Strodthoff, N., & Herwald, H. (2025) · Critical Care · doi:10.1186/s13054-025-05588-0 · Cited by 49
“While several reviews have explored the promise of AI in sepsis management, focus mainly either on early detection models [21, 22] or on specific algorithmic techniques [23, 24], but to a less extend aspects such as clinical translation, regulatory evaluation, and real-world implementation.”
Artificial intelligence and infectious diseases: Scope and perspectives
Abbara, S., Crabol, Y., de Bouillé, J. G., Dinh, A., & Morquin, D. (2025) · Infectious Diseases Now · doi:10.1016/j.idnow.2025.105131 · Cited by 20
Recent approaches of artificial intelligence in intensive care unit: A review
Havelikar, U. A., Bembde, A., Patil, D., Shinde, A., Sharma, R. K., Patel, V. P., & Chaudhari, N. B. (2025) · Intelligent Hospital · doi:10.1016/j.inhs.2025.100030 · Cited by 17
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
“Furthermore, Stylianides et al. [38] present significant advancements in AI applications within intensive care units, emphasizing sepsis prediction.”
Trustworthy AI in digital health: a comprehensive review of robustness and explainability
Mamun, A., Soumma, S. B., & Ghasemzadeh, H. G. (2026) · Progress in Biomedical Engineering · doi:10.1088/2516-1091/ae4e74 · Cited by 11
“Recent reviews further underscore that trustworthy AI in critical care requires not only predictive accuracy but also explainability, ethical safe-guards, and resilience to data variability [171].”
Cha, H., & Kim, J. (2025) · Clinical and Experimental Emergency Medicine · doi:10.15441/ceem.25.199 · Cited by 6
“Several bias detection and mitigation techniques have been proposed, including training AI models with diverse and representative datasets [8,12,13,28] and applying methods that integrate fairness constraints during algorithm development (e.g., debiasing schemes that weight subgroups equally)…”
Zhang, Y., Kirchler, T., & Wang, A. (2026) · Journal of Medical Systems · doi:10.1007/s10916-026-02376-3 · Cited by 7
“It is widely acknowledged in current studies that establish - ing standardized methods is a foundational step towards transparency [30, 74, 81].”
Fairness-Aware Graph Neural Networks for ICU Length of Stay Prediction in IoT-Enabled Environments
Maroudis, A.-C., Karathanasopoulou, K. N., Stylianides, C., Dimitrakopoulos, G. J., & Panayides, A. S. (2025) · IEEE Access · doi:10.1109/access.2025.3560180 · Cited by 4
“…numerous novel works in health and medicine, for instance, GNN-based frameworks have proven effective in fault diagnosis (FD) by capturing intricate dependencies within structured data [15], there is yet only a minor sample of graph-based solutions for LOS prediction and similar challenges [50].”
Self-citationCăliman-Sturdza, O. A., Gheorghita, R. E., Filip, R., & Lobiuc, A. (2026) · Diagnostics · doi:10.3390/diagnostics16121827 · Cited by 5
“The decision-makers ought to incorporate economic and equity-based criteria in procurement and governance to realize the potential of AI [83,84].”
Preprocessing and Visualisation of Dataset for Neonatal Sepsis Prediction Using MIMIC-IV
Ranade, M. D. (2025) · International Conference on Visual Analytics and Data Visualization (ICVADV) · doi:10.1109/icvadv63329.2025.10961922 · Cited by 1
“Stylianides et al. (2025) discussed the uses of AI in intensive care units (ICUs), with an emphasis on sepsis prediction, highlighting deep learning and predictive analytics advancements [9].”
Chunked Data Shapley: A Scalable Dataset Quality Assessment for Machine Learning
Loizou, A., & Tsoumakos, D. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761305 · Cited by 1
Yang, X., Lv, Z., Lv, H., Zhou, Q., Jiang, W., Ziqun, M. E. N. G., & Yang, W. (2026) · JMIR Medical Informatics · doi:10.2196/82762 · Cited by 1
“Timely identification and early intervention are critical for improving patient outcomes, as delays in treatment significantly increase mortality and morbidity [4,5].”
Explainable Artificial Intelligence in Critical Care Nursing: A Discussion Paper
Berdida, D. J. E. (2026) · Nursing in Critical Care · doi:10.1111/nicc.70560 · Cited by 1
“For example, an explainable sepsis alert that highlights rising lactate levels, increasing oxygen requirements, temperature instability, or changes in vital sign variability enables nurses to judge whether the alert aligns with bedside observations and the patient's trajectory [46, 47].”
Challenges of Artificial Intelligence in the Healthcare Industry 4.0
Jain, M., Dhir, S., & Sinha, S. (2025) · Flexible systems management · doi:10.1007/978-981-96-4667-8_5
Designing reliable algorithms to improve patient outcomes by deploying AI in medical domain
Joshi, R., Kumari, S., & Pandey, K. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-41660-6.00019-3
Tsai, T., Hong, C.-C., Cheng, H.-W., & Yang, C.-A. (2025) · Bioinformatics Advances · doi:10.1093/bioadv/vbaf322
“It has been reported that small-size dataset might contribute to overfitting, and the generalizability of the model could be hindered without external validations (Stylianides et al. 2025).”
Shanthi, R., Sharma, M., Khairnar, P. N., Banupriya, R., Rajalakshmi, S., & Jegajothi, B. (2025) · International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT) · doi:10.1109/icerect65215.2025.11376180
RetractedSikkander, A. R. M., Rodrigues, J. J. P. C., Abuelmakarem, H. S., & Meena, M. (2025) · World Journal of Applied Medical Sciences · doi:10.65336/wjams.2025.21103
Alkahazraji, R. (2026) · Acta Colombiana de Cuidado Intensivo · doi:10.1016/j.acci.2026.100609
Mathur, S. (2025) · International Journal of Advances in Signal and Image Sciences · doi:10.29284/zdfexk19
Geetha, A., Nisha, K. S., Sankar, M. S. A., & Rajeev, S. P. (2026) · Machine Learning and Knowledge Extraction · doi:10.3390/make8060150
“Stylianides et al. [15] did a comprehensive review on AI in ICUs with a focus on sepsis prediction and discussed various machine learning architectures that have improved real-time sepsis detection.”
Shakoori, S. R., Asif, M., Fatima, A., Khan, M. A., Murtaza, A., & Ali, T. (2026) · Medical Forum Monthly · doi:10.60110/medforum.370515
Artificial intelligence for early sepsis detection and dynamic prognostication in onco-critical care
Sirohiya, P., Maurya, P., Arora, S., Ratre, B. K., Singh, R., & Kumar, B. (2026) · World Journal of Critical Care Medicine · doi:10.5492/wjccm.120560
“Currently, the vast majority of AI models in onco-critical care exist as retrospective proofs-of-concept[18,25].”
InfluentialExplainable Artificial Intelligence for Early Warning of Sepsis in ICU Patients
Ratul, M. H., Rimti, R. N., Rahman, S., & Nishat, F. I. (2026) · IEEE International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN) · doi:10.1109/qpain69676.2026.11546141
“…to this, and emphasized that explainable AI, multimodal data, Transformer based models, online learning, prospective trials, unified definitions of Sepsis 3 across hospitals, and tight collaboration between AI and clinicians are all necessary to deploy AI in the real world in the ICU [11].”
Keshavarz, Z., Zand, F., Asmarian, N., Rafiee, M., & Sarabi, R. E. (2026) · BMJ Open · doi:10.1136/bmjopen-2025-109036
Combining machine learning and physiological network models for sepsis prediction
Backes, J., Tsanda, A., Knopp, T., Renz, W., & Schöll, E. (2026) · Frontiers in Network Physiology · doi:10.3389/fnetp.2026.1852577
“Second, the explainability of these methods predominantly relies on Shapley-value analyses, deriving importance factors of single input features or input-feature interactions (Stylianides et al., 2025; Sundararajan and Najmi, 2020).”
Garcha, G., Scott, M., Rao, S., & Rao, R. (2026) · INNOVAPATH · doi:10.63501/pj8sbm22
“AI has emerged as a powerful tool for early sepsis detection, with various algorithms demonstrating the ability to predict sepsis hours before clinical manifestation (Papaioannou et al ., 2025).”
Explainable Mortality Outcome Prediction Using Neuro-Symbolic Causal Reasoning
Bhogawar, T., Panchal, R., Kabra, A., Chauhan, N., & Sweta, S. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-31233-4_19
The benefits and biases of artificial intelligence in microbiology
Aransiola, S. A., Onyinoyi, A. H., Akhigbe, J. O., Onoruoiza, S. I., Esenogho, E., & Maddela, N. R. (2026) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2026.109457
Interpretable Machine Learning for Early Sepsis Prediction
Stylianides, C., Nicolaou, A., Vavlitou, A., Palazis, L., Kyriacou, E. C., Pattichis, M. S., Pattichis, C. S., & Panayides, A. S. (2026) · Proceedings - IEEE Symposium on Computer-Based Medical Systems · doi:10.1109/cbms69103.2026.00062
Self-citationKusuma, S., Karthik Reddy, M., & Kiran Kumar, M. (2026) · International Research Journal of Multidisciplinary Scope · doi:10.47857/irjms.2026.v07i04.010543
“Moreover, due to variability of disease presentations in sepsis, rule-based score systems lack both sensitivity and specificity, thus requiring the development of a more adaptive data-driven solution (7).”
Metadata
- Authors
- Stylianides, C., Nicolaou, A., Sulaiman, W. A., Alexandropoulou, C.-A., Panagiotopoulos, I., Karathanasopoulou, K., Dimitrakopoulos, G., Kleanthous, S., Politi, E., Ntalaperas, D., Papageorgiou, X., Garcia, F., Antoniou, Z., Ioannides, N., Palazis, L., Vavlitou, A., Pattichis, M. S., Pattichis, C. S., & Panayides, A. S.
- Affiliations
- Centre of Excellence, CYENS, Nicosia 1016, Cyprus
- Harokopio University of Athens (GR)
- Department of Informatics and Telematics, Harokopio University of Athens, 176 76 Kallithea, Greece
- Open University of Cyprus (CY)
- UBITECH Limited, Limassol 3071, Cyprus
- Research & Development Department, 3aHealth, Strovolos 2020, Cyprus
- State Health Services Organization, Aglantzia 2100, Cyprus
- University of New Mexico (US)
- Corresponding author
- Stylianides, Charithea
- ORCID
- Venue
- Machine Learning and Knowledge Extraction
- Publisher
- MDPI AG
- Published
- 2025-01-08
- Volume
- 7
- Issue
- 1
- Pages
- 6
- ISSN
- 2504-4990 (ISSN-L)
- OpenAlex
- W4406170188
- Semantic Scholar
- fcfd620a9a6093506eb00d2d7771e830231b33ae
- Cited by
- 28
- References
- 237
- Citations by year
- 2026: 16 · 2025: 12
- FWCI
- 19.52
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Machine Learning and Knowledge Extraction (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Republic of Cyprus (CODEVELOP-ICTHEALTH/0322/0071)
- 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 › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- sepsis prediction, intensive care unit, artificial intelligence in healthcare, sepsis mortality prediction, hospital length of stay prediction, ICU admission prediction, emergency department admission, diagnostic biomarkers, explainable AI, healthcare resource allocation, multimodal data, transformer models
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.