Radiomics and anaesthetic planning: Quantitative imaging as a new frontier in preoperative risk assessment

Radiomics-the high-throughput extraction of quantitative features from standard medical images-has transformed oncologic imaging by revealing subvisual patterns linked to tissue biology, yet its role in perioperative medicine remains largely unexplored. This narrative review summarises current evidence linking imaging-derived radiomic biomarkers to perioperative outcomes and proposes a conceptual framework for integrating radiomics into precision anaesthesia. Quantitative assessment of body composition, organ function, and vascular morphology from routine preoperative computed tomography and magnetic resonance imaging can provide objective indicators of physiologic reserve, drug-handling capacity, and recovery potential. Across heterogeneous, largely oncological cohorts, combined radiomic-clinical models have reported higher discrimination (area under the curve 0.84-0.93) than conventional risk scores for selected postoperative complications, and artificial intelligence-based airway assessment has shown sensitivity and specificity exceeding traditional bedside tests; these figures are pooled from methodologically diverse studies rather than single validated estimates. The Image Biomarker Standardisation Initiative has substantially reduced cross-platform feature variability. However, perioperative-specific, prospectively validated evidence remains scarce. Prospective multicentre trials, standardized and automated feature-extraction pipelines, transparent cost and equity appraisal, and integration with electronic health records are critical priorities before radiomics-driven preoperative assessment can enter routine anaesthetic practice.