MACHINE LEARNING MODELS FOR PREDICTING PATIENT OUTCOMES IN INTENSIVE CARE UNITS: A CASE STUDY IN U.S. HOSPITALS
DOI:
https://doi.org/10.48047/y0t69v56Keywords:
Intensive Care Unit, Machine Learning, Outcome Prediction, Interpretability, XGBoost.Abstract
The integration of machine learning (ML) techniques into intensive care unit (ICU) settings has opened new avenues for predicting critical patient outcomes, thereby enabling timely interventions and optimized clinical resource utilization. This study presents the development and evaluation of several supervised ML models aimed at predicting two vital ICU outcomes—length of stay (LOS) and in-hospital mortality—based on data collected within the first 24 hours of ICU admission. Leveraging the large-scale, publicly available MIMIC-III database, the study utilizes a diverse set of clinical variables, including demographics, vital signs, laboratory values, and administrative records, to construct robust predictive frameworks. The models explored include logistic regression, random forests, XGBoost, and feedforward neural networks, with logistic regression emerging as the top performer. For the classification task, distinguishing short (≤4 days) and long (>4 days) ICU stays, the logistic regression model achieved an accuracy of 88.0%, an AUC-ROC of 0.876, and an F1-score of 84.6%. Random Forest demonstrated comparable performance with identical accuracy (88.0%) and precision (89.2%), achieving an AUC-ROC of 0.855 and F1-score of 84.6%. XGBoost showed slightly lower but still robust performance with an accuracy of 87.0%, precision of 86.8%, F1-score of 83.5%, and AUC-ROC of 0.875. The neural network model yielded the lowest performance among the four approaches, achieving an accuracy of 83.0% and an AUC-ROC of 0.862. In addition to performance optimization, the study integrates explainability through SHAP and LIME, offering insights into feature contributions and supporting transparent, clinician-friendly model interpretation. The inclusion of fairness assessments further strengthens the ethical integrity of these models by addressing bias across age, gender, and ethnicity. This research underscores the feasibility and importance of incorporating interpretable, generalizable, and ethically aware machine learning models into ICU decision-support systems. Ultimately, the findings contribute to the advancement of AI-driven precision medicine, with the potential to transform care delivery in critical settings by enhancing early risk stratification, improving outcome prediction, and supporting evidence-based clinical decisions.
Downloads
References
Ahmad, J., F. Tauseef, and Z. Akbar, Predictive analytics for AI-assisted patient no-show management and clinic revenue optimization: a simulation-based research. Migration Letters, 2024. 21(S13): p. 1901-1924.
Aziz, A.S., et al. A Comprehensive Framework Analysis of Cycle GAN-Based Modality Translation: Enhancing Brain Tumor Diagnostics from FLAIR to T2w. in 2025 4th International Conference on Electronics Representation and Algorithm (ICERA). 2025. IEEE.
Sharif, K.S., M.M. Uddin, and M. Abubakkar. Neurosignal precision: A hierarchical approach for enhanced insights in parkinson's disease classification. in 2024 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA). 2024. IEEE.
Knaus, W.A., et al., The APACHE III prognostic system: risk prediction of hospital mortality for critically III hospitalized adults. Chest, 1991. 100(6): p. 1619-1636.
Siyam, O.F., et al., Interpretable Deep Learning for Symptom-Based Lung Cancer Prediction Using a 1D CNN Framework.
Vincent, J.-L., et al., The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure: On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine (see contributors to the project in the appendix). Intensive care medicine, 1996. 22(7): p. 707-710.
Johnson, A.E., et al., MIMIC-III, a freely accessible critical care database. Scientific data, 2016. 3(1): p. 1-9.
Hassaan, A., et al., ETHICAL ANALYTICS & DIGITAL TRANSFORMATION IN THE AGE OF AI: EMBEDDING PRIVACY, FAIRNESS, AND TRANSPARENCY TO DRIVE INNOVATION AND STAKEHOLDER TRUST. Contemporary Journal of Social Science Review, 2023. 1(04): p. 1-18.
Jamshaid, M.M., et al., Preparing human oversight talent for agentic AI workplaces: A competency framework for education and workforce systems. Contemporary Journal of Social Science Review, 2025. 3(4): p. 1561-1574.
Komorowski, M., et al., The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive care. Nature medicine, 2018. 24(11): p. 1716-1720.
Sharif, K.S., et al. DNA Sequence Classification: An Advanced Machine Learning Framework For Accurate Splice Junction Detection. in 2025 International Conference on Metaverse and Current Trends in Computing (ICMCTC). 2025. IEEE.
Shickel, B., et al., Deep EHR: a survey of recent advances in deep learning techniques for electronic health record (EHR) analysis. IEEE journal of biomedical and health informatics, 2017. 22(5): p. 1589-1604.
Harutyunyan, H., et al., Multitask learning and benchmarking with clinical time series data. Scientific data, 2019. 6(1): p. 96.
Saeed, M., Data-Driven Healthcare: The Role of Business Intelligence Tools in Optimizing Clinical and Operational Performance. The American Journal of Applied Sciences, 2025. 7(8): p. 50-73.
Lundberg, S.M. and S.-I. Lee, A unified approach to interpreting model predictions. Advances in neural information processing systems, 2017. 30.
Uddin, M.M., et al. Interpretable Alzheimer's Disease Diagnosis Via CNNs and MRI: an Explainable AI Approach. in 2025 4th International Conference on Electronics Representation and Algorithm (ICERA). 2025. IEEE.
Pollard, T.J., et al., The eICU Collaborative Research Database, a freely available multi-center database for critical care research. Scientific data, 2018. 5(1): p. 180178.
Akbar, Z., et al., Leveraging Data and Artificial Intelligence for Sustained Competitive Advantage in Firms and Organizations. Journal of Innovative Computing and Emerging Technologies, 2023. 3(1).
Mahmud, T., AUTONOMOUS BOUNDARY ALERT SYSTEM FOR CHILD/CRIMINAL MONITORING. 2020. DAFFODIL INTERNATIONAL UNIVERSITY.
Goldberger, A.L., et al., PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. circulation, 2000. 101(23): p. e215-e220.
Lipton, Z.C., et al., Learning to diagnose with LSTM recurrent neural networks. arXiv preprint arXiv:1511.03677, 2015.
Ahmed, M.S., et al., Ai-driven predictive operations management: a business science framework for dynamic hospital resource optimization and clinical workflow efficiency. International Journal of Professional Business Review: Int. J. Prof. Bus. Rev., 2025. 10(8): p. 5.
Sharif, K.S., et al. A comparative framework integrating hybrid convolutional and unified graph neural networks for accurate parkinson’s disease classification. in 2024 7th International Seminar on Research of Information Technology and Intelligent Systems (ISRITI). 2024. IEEE.
Rajkomar, A., et al., Ensuring fairness in machine learning to advance health equity. Annals of internal medicine, 2018. 169(12): p. 866-872.
Niaz, S., et al., AI for Inclusive Educational Governance and Digital Equity Examining the Impact of AI Adoption and Open Data on Community Trust and Policy Effectiveness. Contemporary Journal of Social Science Review, 2024. 2(04): p. 2557-2567.
Zamil, M.Z.H., et al. Stroke Prediction on Healthcare Data Using SMOTE and Explainable Machine Learning. in 2025 13th International Symposium on Digital Forensics and Security (ISDFS). 2025. IEEE.
Arafat, Y., et al., Big Data Analytics in Information Systems Research: Current Landscape and Future Prospects Focus: Data science, cloud platforms, real-time analytics in IS. Emerging Frontiers Library for The American Journal of Engineering and Technology, 2025. 7(8): p. 177-201.
Mondal, K.K., M.S. Uddin, and T. Mahmud, From Pandemic to Persistence: Obesity, Immunity, and Next-Gen COVID-19 Vaccines. 2025.
Meyer, A., et al., Machine learning for real-time prediction of complications in critical care: a retrospective study. The Lancet Respiratory Medicine, 2018. 6(12): p. 905-914.
Begum, S., et al., AttenGene: A deep learning model for gene selection in PDAC classification using autoencoder and attention mechanism for precision oncology. Well Testing Journal, 2025. 34(S3): p. 705-726.
Abubakkar, M., et al., AI in Neurology: Speech-Based Detection of Parkinson’s Disease using Machine Learning Models. In 2025 IEEE 9th Forum on Research and Technologies for Society and Industry (RTSI)(pp. 152-157). Sweeney, L.(2002). k-anonymity: A model for protecting privacy.* International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 2025. 10(05): p. 557-570.
Rahman, M.B., et al., Appraising the historical and projected spatiotemporal changes in the heat index in Bangladesh. Theoretical and Applied climatology, 2021. 146(1-2): p. 125.
Vellido, A., et al., Machine learning in critical care: state-of-the-art and a sepsis case study. Biomedical engineering online, 2018. 17(Suppl 1): p. 135.
Meiring, C., et al., Optimal intensive care outcome prediction over time using machine learning. PloS one, 2018. 13(11): p. e0206862.
Abubakkar, M., et al. Explainable suicide risk prediction with DeepFusion: A hybrid intelligence approach. in 2025 4th International Conference on Electronics Representation and Algorithm (ICERA). 2025. IEEE.
Rahman, M.B., et al., Enhancing Healthcare Outcomes Through Data-Driven Decision Making: A Business Analytics Approach. Nvpubhouse Library for International Journal of Medical Science and Public Health Research, 2025. 6(10): p. 26-53.
Fernandes, M., et al., Predicting Intensive Care Unit admission among patients presenting to the emergency department using machine learning and natural language processing. PloS one, 2020. 15(3): p. e0229331.
Alghatani, K., et al., Predicting intensive care unit length of stay and mortality using patient vital signs: machine learning model development and validation. JMIR medical informatics, 2021. 9(5): p. e21347.
Roy, K.K., et al., Leveraging artificial intelligence for strategic decision-making in healthcare organizations: a business it perspective. The American Journal of Applied Sciences, 2025. 7(8): p. 74-93.
Rojas, J.C., et al., Predicting intensive care unit readmission with machine learning using electronic health record data. Annals of the American Thoracic Society, 2018. 15(7): p. 846-853.
Abubakkar, M., et al., PROGNOSTIC MODELING FOR HEPATIC DISORDERS: A PARADIGM OF EQUILIBRATED AND GENERALIZED MACHINE LEARNING METHODOLOGIES. American Journal of Scholarly Research and Innovation, 2025. 4(01): p. 352-362.
Sozib, H.M., et al., Cloud Computing in Business: Leveraging SaaS, IaaS, and PaaS for Growth. Journal of Applied Research: p. 38.
Rahman, M.B., et al., Cost-Effective Healthcare Operations: Financial Modeling and Optimization Using Business Intelligence Tools. Nvpubhouse Library for International Journal of Medical Science and Public Health Research, 2025. 6(10): p. 80-106.
Asad, S.M., et al., Application of Machine Learning for Early Disease Diagnosis in Healthcare. Cuestiones de Fisioterapia, 2022. 51(3): p. 332-355.
Cheng, F.-Y., et al., Using machine learning to predict ICU transfer in hospitalized COVID-19 patients. Journal of clinical medicine, 2020. 9(6): p. 1668.
Kanojiya, S., et al., Business Analytics for Quality Improvement: A Case Study in Healthcare Systems. Nvpubhouse Library for International Journal of Medical Science and Public Health Research, 2025. 6(10): p. 137-162.
Khajehali, N., Z. Khajehali, and M.J. Tarokh, The prediction of mortality influential variables in an intensive care unit: a case study. Personal and Ubiquitous Computing, 2023. 27(2): p. 203-219.
Khatun, M. and M.S. Oyshi, Advanced Machine Learning Techniques for Cybersecurity: Enhancing Threat Detection in US Firms. Journal of Computer Science and Technology Studies, 2025. 7(2): p. 305-315.
Greco, M., et al., Outcome prediction during an ICU surge using a purely data-driven approach: A supervised machine learning case-study in critically ill patients from COVID-19 Lombardy outbreak. International journal of medical informatics, 2022. 164: p. 104807.
Fenn, A., et al., Development and validation of machine learning models to predict admission from emergency department to inpatient and intensive care units. Annals of emergency medicine, 2021. 78(2): p. 290-302.
Taylor, R.A., et al., Prediction of in‐hospital mortality in emergency department patients with sepsis: a local big data–driven, machine learning approach. Academic emergency medicine, 2016. 23(3): p. 269-278.
Hasan, M., et al., Predictive Analytics in Cancer Care: Leveraging Machine Learning and Big Data for Early Detection and Treatment Optimization. Nvpubhouse Library for International Journal of Medical Science and Public Health Research, 2025. 6(10): p. 54-79.
Rahman, M.B., et al., Heart Disease Risk Prediction Using Machine Learning and Explainable Ai: A Data-Driven Approach for Early Diagnosis and Prevention.
Desautels, T., et al., Prediction of early unplanned intensive care unit readmission in a UK tertiary care hospital: a cross-sectional machine learning approach. BMJ open, 2017. 7(9): p. e017199.
Nemati, S., et al., An interpretable machine learning model for accurate prediction of sepsis in the ICU. Critical care medicine, 2018. 46(4): p. 547-553.
Kong, G., K. Lin, and Y. Hu, Using machine learning methods to predict in-hospital mortality of sepsis patients in the ICU. BMC medical informatics and decision making, 2020. 20(1): p. 251.
Kuhn, M. and K. Johnson, Applied predictive modeling. Vol. 26. 2013: Springer.
Teasdale, G. and B. Jennett, Assessment of coma and impaired consciousness: a practical scale. The lancet, 1974. 304(7872): p. 81-84.
Chen, T. and C. Guestrin. Xgboost: A scalable tree boosting system. in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016.
Heaton, J., Ian goodfellow, yoshua bengio, and aaron courville: Deep learning: The mit press, 2016, 800 pp, isbn: 0262035618. Genetic programming and evolvable machines, 2018. 19(1): p. 305-307.
Bergstra, J. and Y. Bengio, Random search for hyper-parameter optimization. Journal of machine learning research, 2012. 13(2).
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation .
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
