Impact of Artificial Intelligence on Radiologist Workflow and Diagnostic Accuracy in Tokyo Hospitals: A Comparative Study

Abstract

Introduction: The integration of Artificial Intelligence (AI) into radiology holds the promise of enhancing workflow efficiency and diagnostic accuracy. This study aimed to evaluate the impact of AI implementation on radiologist workflow and diagnostic performance in Tokyo hospitals.

Methods: A comparative study was conducted across five major hospitals in Tokyo. Two hospitals had fully integrated AI into their radiology workflow, while three served as controls with conventional practices. Data were collected on reading time, diagnostic accuracy, and radiologist satisfaction through surveys and performance metrics. Statistical analysis was performed to compare AI-integrated and control groups.

Results: AI integration was associated with a significant reduction in reading time per case (p < 0.05). Diagnostic accuracy improved modestly in complex cases with AI assistance (p < 0.05). Radiologist satisfaction surveys indicated increased confidence and reduced burnout in the AI-integrated group.

Conclusion: AI implementation in Tokyo hospitals positively impacts radiologist workflow by decreasing reading time. While improvements in diagnostic accuracy are notable in complex cases, the overall impact is modest. AI contributes to enhanced radiologist satisfaction. Further research is warranted to explore long-term effects and address potential challenges.