Perioperative management is essential for ensuring surgical safety and improving patient outcomes. With increasing surgical complexity and patient heterogeneity, traditional experience-based decision-making is no longer sufficient to meet the demands of precision medicine. Recent advances in artificial intelligence (AI), big data analytics, and large language models (LLMs) provide new opportunities to transform perioperative clinical decision support. AI-driven models enable accurate risk stratification and prediction of perioperative complications, while big data technologies facilitate multimodal data integration, real-time monitoring, and personalized intervention strategies. Meanwhile, LLMs enhance clinical communication, support medical documentation, and assist knowledge-based clinical decision-making through advanced natural language processing capabilities. However, despite rapid technological development, a comprehensive framework integrating AI, big data, and LLMs for perioperative intelligent decision-making remains insufficiently explored. This article reviews emerging applications of AI in preoperative risk assessment, intelligent anesthesia management, and postoperative complication prediction, and summarizes the role of big data in integrated clinical platforms and personalized treatment strategies. It also highlights the potential of  LLMs in patient education, clinical decision support systems, and automated knowledge synthesis. Overall, integrating AI, big data, and LLMs may establish an interpretable closed-loop perioperative decision-support ecosystem that improves surgical safety, clinical efficiency, and personalized healthcare delivery.

Sepsis-related myocardial dysfunction significantly increases patient mortality and remains a major challenge in critical care medicine. Its pathological mechanisms involve uncontrolled inflammation, excessive oxidative stress, and impaired stress tolerance of cardiomyocytes. Currently, no specific treatment methods are available. Although supportive treatment remains the mainstream in clinical practice, effective interventions for septic myocardial injury are still lacking. This review systematically summarizes the cardioprotective effects of traditional Chinese medicine (TCM) preparations, including Xuebijing injection (containing safflower, red peony root, and salvia miltiorrhiza) and Shenfu Injection (containing ginseng and aconite), in cell and animal models. The mechanisms include antioxidation, anti-inflammation, and enhancement of cardiomyocyte resistance to injury. Some of these preparations have been recommended in the "Guidelines for the Diagnosis and Treatment of Sepsis" in China and are used as adjunctive treatments in intensive care units (ICUs). Preliminary clinical studies suggest  that these agents may improve microcirculation and potentially enhance survival rates. This review not only highlights the multi-target potential of TCM in sepsis-induced myocardial injury but also provides a direction for future high-quality mechanistic research and international multicenter randomized controlled trials, thereby facilitating the integration of traditional medicine into modern critical care practice.

Minimally invasive surgeries are widely applied due to the advantages of small surgical wound, short postoperative recovery period, and low surgical infection rate. In minimally invasive surgery, vascular clips play a significant role in hemostasis and managing the direction of blood flow, ensuring the success of surgical procedures. The mechanical structure and manufacturing materials of vascular clips have an important influence on its clinical application effect. In this paper, we classify and summarize the mechanical structure and manufacturing materials of the currently available vascular clips worldwide, and then analyze the hemostatic coating materials of vascular clips. Besides, we also summarize the shortcomings of the existing vascular clips and propose a coat-absorbable vascular clip with future research potential.

The mandible plays a crucial role in facial structure and function. Mandibular defects, often caused by tumor resection, trauma, or congenital conditions, severely impact patients' physiological functions and quality of life. Reconstruction methods for mandibular defects primarily include autologous bone grafting, computer-aided reconstruction, and distraction osteogenesis. Among these, autologous bone grafting remains the most widely used technique. As medical technologies advance, the integration of computer-aided design and 3D printing has provided new technical support for personalized mandibular reconstruction, improving both surgical precision and safety. Moreover, distraction osteogenesis has emerged as a promising alternative in recent years. This paper reviews research on fixation methods for V mandibular defect reconstruction to offer clinical guidance for effective mandibular repair.