Machine Learning and Artificial Intelligence in Intensive Care Medicine: Critical Recalibrations from Rule-Based Systems to Frontier Models

Abstract

Artificial intelligence (AI) and machine learning (ML) are rapidly transforming clinical decision support systems (CDSSs) in intensive care units (ICUs), where vast amounts of real-time data present both an opportunity and a challenge for timely clinical decision-making. Here, we trace the evolution of machine intelligence in critical care. This technology has been applied across key ICU domains such as early warning systems, sepsis management, mechanical ventilation, and diagnostic support. We highlight a transition from rule-based systems to more sophisticated machine learning approaches, including emerging frontier models. While these tools demonstrate strong potential to improve predictive performance and workflow efficiency, their implementation remains constrained by concerns around transparency, workflow integration, bias, and regulatory challenges. Ensuring the safe, effective, and ethical use of AI in intensive care will depend on validated, human-centered systems supported by transdisciplinary collaboration, technological literacy, prospective evaluation, and continuous monitoring.

Keywords: artificial intelligence; clinical decision support systems; critical care; deep learning; intensive care medicine; large language models; machine learning; natural language processing; predictive analytics; privacy; safety; sepsis.

Figure 1

Key technological milestones in AI and ML applications for biomedicine and intensive care medicine. The upper timeline (blue) depicts landmark developments in general biomedical applications, spanning from early theoretical foundations to current specialized systems as shown in Supplementary Table S1. The lower timeline (purple) illustrates parallel innovations specific to intensive care medicine, highlighting progression from basic clinical decision support to sophisticated predictive analytics, as shown in Supplementary Table S2. A concise description of each timepoint, with corresponding scientific references, is provided in the Supplementary Material. Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; CAD, computer-aided diagnosis; ELIZA, early natural language processing system; HELP, Health Evaluation through Logical Processing; MIMIC, Medical Information Mart for Intensive Care; SICULA, Super ICU Learner Algorithm; SVM, support vector machine; TREWScore, Targeted Real-time Early Warning Score.

Figure 2

Mind-map illustrating key aspects of AI/ML integration in ICUs.