AI-Driven Clinical Decision Support for Personalized Musculoskeletal Care: A Retrieval-Augmented Generation Approach

Artificial intelligence ( AI) demonstrates significant promise for transforming healthcare, especially in diagnosing and managing musculoskeletal disorders, which enhances patient interactions. This paper introduces a novel retrieval augmented generation (RAG) methodology to develop an AI-driven clinical decision support system to improve the accuracy of diagnostic decision-making and personalized treatment planning. The proposed RAG model leverages a BERT-based medical text classification algorithm for initial assessment and then retrieves similar historical cases to generate tailored treatment plans and medication recommendations using BioGPT. A very important aspect of the workflow includes physician validation of the AI-generated information before it is shown to the patient, ensuring safety and clinical oversight. The system’s performance showed excellent results among key retrieval metrics, which includes precision (0.87), recall (0.85), and F1 score (0.86), showing the potential of AI systems to effectively support and improve remote healthcare in orthopedic and physical therapy settings.