Venipuncture Via Nurse-AI Collaboration

The goal of this clinical trial is to learn if AI-assisted site selection plus human-machine coordinated puncture improves the quality and safety of peripheral venous access for contrast-enhanced imaging in patients of all ages needing CT/MR contrast scans across three hospitals. The main questions it aims to answer are: Does the AI-assisted workflow improve the composite success rate of venous cannulation compared with routine nursing? Does this human-machine model raise first-attempt puncture success, shorten procedure time and reduce complications? Researchers will compare the AI-assisted intervention group to the routine nursing control group to see if the AI-assisted human-machine coordinated puncture brings clinical benefits. Participants will: * Receive peripheral venous cannulation for contrast-enhanced imaging. * Have arm images and clinical data collected before and after puncture. * Complete standardized venous access assessment and 24-hour post-procedure safety observation.

Inclusion Criteria: * Patients of all ages scheduled for contrast-enhanced CT/MR examination at three participating centers, who require peripheral venous access. * Recent renal function test showing estimated glomerular filtration rate (eGFR) \> 30 ml/min/1.73 m². * Written informed consent obtained from the patient or legal representative, agreeing to radiology specialized nursing service. Exclusion Criteria: * Patients with irreversible brain injury or brain death. * Contraindications to peripheral venipuncture, including history of bilateral mastectomy, extremity venous thrombosis, etc. * eGFR ≤ 30 ml/min/1.73 m². * Uncontrolled hyperthyroidism. * Patients planned for Iodine-131 therapy within the next two weeks. * Extremely poor vascular conditions that make indwelling needle cannulation impossible. * Patients receiving contrast-enhanced imaging of both upper limbs. * Known contraindications to contrast media.