From Accuracy to Adoption: A Clinical Readiness Framework for Artificial Intelligence in Orthopaedic Surgery (OSCAR)

Technical performance alone does not establish clinical readiness; strong accuracy or discrimination does not demonstrate that an artificial intelligence (AI) tool will generalize, improve clinical care, or remain safe after deployment. Existing AI frameworks provide important guidance for reporting, risk of bias, model evaluation, and clinical trials, but they do not provide a single longitudinal stage describing a technology's progression from development through monitored clinical deployment. We propose the Orthopaedic Surgery Clinical AI Readiness (OSCAR) scale as a 6-stage conceptual framework spanning model development, internal and external validation, early clinical evaluation, comparative clinical validation, and monitored deployment. Illustrative applications of OSCAR demonstrate how technically strong models may remain at stage 1, external validation advances a tool to stage 2, live workflow evaluation represents stage 3, and comparative clinical evidence is required for stage 4. OSCAR is intended to complement, not replace, established reporting and risk-of-bias standards. The framework remains unvalidated and requires stakeholder consensus, explicit decision criteria, inter-rater reliability testing, and prospective evaluation.