From static to dynamic: Artificial intelligence revolution in perioperative care through multimodal data fusion and closed-loop optimization

Mingdi Xue; Jiaming Yang; Honglin Wang; Zineng Yan; Xiaoliang Chen; Weihang Gao; Rui Luo; Xiangxin Lv; Zhewei Ye

doi:10.1016/j.jatmed.2025.06.003

From static to dynamic: Artificial intelligence revolution in perioperative care through multimodal data fusion and closed-loop optimization

J Anesth Transl Med. 2025 Aug 13;4(3):132-141. doi: 10.1016/j.jatmed.2025.06.003. eCollection 2025 Sep.

Authors

Mingdi Xue^{1

2}, Jiaming Yang^{1

2}, Honglin Wang^{1

2}, Zineng Yan^{1

2}, Xiaoliang Chen^{1

2}, Weihang Gao^{1

2}, Rui Luo^{1

2}, Xiangxin Lv^{1

2}, Zhewei Ye^{1

2}

Affiliations

¹ Department of Orthopedics, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430000, China.
² Intelligent Medical Laboratory, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430000, China.

Abstract

The integration of artificial intelligence (AI) into perioperative care represents a paradigmatic transformation from static decision frameworks to dynamic, adaptive systems. This comprehensive review synthesizes recent breakthroughs in multimodal data fusion and closed-loop optimization for perioperative care, examining the convergence of technological innovation with clinical imperatives. We systematically analyze three critical dimensions: the fundamental paradigm shift driven by clinical complexity, technological breakthroughs in multimodal fusion architectures, and dynamic decision-making implementations in anesthesia depth regulation and pain management optimization. Our analysis reveals that contemporary perioperative environments demand sophisticated AI systems capable of real-time data integration and dynamic response optimization. We identify critical bottlenecks in clinical implementation, including computational optimization under real-time constraints, robustness assurance mechanisms, and interpretability challenges. Looking forward, we delineate three interconnected pillars essential for advancing the field: standardized multimodal benchmark datasets, human-AI collaborative decision-making paradigms, and ethical-regulatory frameworks. This review establishes a comprehensive roadmap for transforming perioperative care through AI integration, emphasizing the synergistic relationship between technological innovation and clinical expertise in optimizing patient outcomes.

Keywords: Artificial intelligence; Closed-loop optimization; Multimodal data fusion; Perioperative care.

Publication types

Review