A174: Visualization Analysis of Artificial Intelligence Applications in Sports

Artificial intelligence (AI) has revolutionized sports science, advancing performance analysis, injury prevention, and strategy optimization. However, its long-term impact remains underexplored. This study conducts a bibliometric and predictive analysis of -driven sports research over the past decade, identifying key contributors, emerging trends, and future directions through visualization techniques. Method: A systematic review was conducted on related sports research from 2014 to 2024 using the Web of Science Core Collection. Bibliometric tools Citespace and Vosviewer were employed to analyze publication trends, author networks, institutional collaborations, and keyword co-occurrences. Polynomial regression analysis was applied to forecast future research growth based on historical publication and citation trends. A total of 5,811 publications with 96,753 citations were identified. China, the United States, and the United Kingdom were the most productive countries, with China leading in volume but exhibiting lower citation impact. The Chinese Academy of Sciences, Stanford University, and the University of Oxford were the top research institutions. Keyword analysis revealed that "machine learning," "deep learning," and "computer vision" were the most studied topics, while emerging themes such as "stress analysis," "information processing," and "pose estimation" indicated shifts towards driven real-time monitoring and predictive analytics. Polynomial regression models predicted continued research growth, with publication trends following y = 354x ² + 1350x + 528 (r ² = 0.94) and citation growth modeled as y = 9680x ² + 24900x + 7850 (r ² = 0.99), suggesting sustained acceleration in AI applications within sports science. The integration of sports science has grown rapidly, with machine learning and computer vision playing a pivotal role in optimizing athletic performance, real-time feedback, and injury prevention. Predictive analytics and driven modeling can transform sports training by personalizing programs based on biomechanical and physiological data, enhancing both performance and injury resilience. However, disparities in AI adoption across regions and institutions highlight the need for greater international collaboration, particularly in developing regions where access to driven sports technologies is limited. Future research should refine -driven models for individualized training, integrate wearable sensor data for precision, and address ethical concerns. Additionally, policymakers and sports organizations should invest in AI-based training and health monitoring systems to bridge the gap between technology-rich and technology-limited regions. As it evolves, its role in sports science will expand, driving advancements in performance analysis, health monitoring, and strategic decision-making.