In emergency medical situations within ambulances, medical personnel often face challenges such as time constraints, limited equipment, confined spaces, and unstable network connections. These conditions are exacerbated when complex surgical procedures are required, and emergency responders frequently lack the specialized knowledge and experience needed, which can significantly impact patient survival rates and treatment outcomes. To address these issues, this paper proposes a surgical assistance system based on augmented reality (AR), integrating AR technology with edge computing, cloud computing, and artificial intelligence (AI), specifically designed for emergency medical care in ambulances. The system offers real-time 3D surgical tutorials, dynamic error detection and feedback, cloud-based tutorial retrieval, and remote guidance with AI learning functions. It aims to help emergency personnel make quick and accurate decisions in unfamiliar surgical scenarios, thereby enhancing the safety and effectiveness of emergency procedures and ultimately improving patient outcomes. CSCI-RTHI