Artificial intelligence-assisted medical coding and DRG management: current applications, challenges, and future perspectives

Abstract

Artificial intelligence (AI) is reshaping the way medical information is processed and has shown considerable potential in medical record coding and diagnosis-related group (DRG) management. Traditional medical record management mainly relies on manual coding and rule-based matching, which is often limited by low efficiency, heavy workload, and insufficient consistency, making it difficult to meet the demands of large-scale healthcare data processing. In recent years, machine learning, deep learning, and natural language processing (NLP) have been increasingly applied to automated coding, clinical information extraction, and medical record quality control, and have gradually expanded to DRG grouping prediction, risk control, and hospital operation management. Large language models (LLMs) offer new opportunities for complex medical text understanding and candidate decision support; however, their application in medical record coding and DRG management remains at an early stage of validation and exploration. At present, this field still faces several challenges, including data heterogeneity, limited model interpretability, privacy and security concerns, and insufficient cross-institutional generalizability. Future efforts should focus on standardized validation, multimodal data integration, and human–AI collaboration mechanisms to promote the robust development of intelligent medical record management systems.