Zhao Bin, Liu Hua, Liu Qiuli, Qi Wenwen, Zhang Weiwen, Du Jianer, Jin Yi, Weng Xiaojian
Department of Anesthesiology and SICU, Xinhua Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China.
Department of Anesthesiology, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University, Shanghai, China.
J Infect Dis. 2025 Feb 20;231(2):e345-e354. doi: 10.1093/infdis/jiae403.
Surgical site infection (SSI) is a common and costly complication in spinal surgery. Identifying risk factors and preventive strategies is crucial for reducing SSIs. Generative Pre-trained Transformer 4 (GPT-4) has evolved from a simple text-based tool to a sophisticated multimodal data expert, invaluable for clinicians. This study explored GPT-4's applications in SSI management across various clinical scenarios.
GPT-4 was employed in clinical scenarios related to SSIs in spinal surgery. Researchers designed specific questions for GPT-4 to generate tailored responses. Six evaluators assessed the responses for logic and accuracy using a 5-point Likert scale. Interrater consistency was measured with Fleiss' kappa, and radar charts visualized GPT-4's performance.
Interrater consistency, measured by Fleiss' kappa, ranged from 0.62 to 0.83. The average scores for logic and accuracy were 24.27 ± 0.4 and 24.46 ± 0.25. Radar charts indicated consistently high performance across criteria. GPT-4 demonstrated proficiency in creating personalized treatment plans, improving SSI management strategies, and identified emerging research trends.
GPT-4 shows a significant potential in SSI management in spinal surgery, promoting patient-centered care and precision medicine. Despite limitations in antibiotics and patient education, GPT-4's continuous learning, data privacy focus, and professional collaboration indicate its potential to revolutionize SSI management, requiring further development.
手术部位感染（SSI）是脊柱手术中常见且代价高昂的并发症。识别风险因素和预防策略对于减少手术部位感染至关重要。生成式预训练变换器4（GPT-4）已从一个简单的基于文本的工具发展成为一个复杂的多模态数据专家，对临床医生来说非常宝贵。本研究探讨了GPT-4在各种临床场景下手术部位感染管理中的应用。
GPT-4被应用于脊柱手术中与手术部位感染相关的临床场景。研究人员为GPT-4设计了特定问题，以生成量身定制的回答。六位评估者使用5点李克特量表评估回答的逻辑性和准确性。使用Fleiss卡方测量评分者间的一致性，并通过雷达图直观展示GPT-4的表现。
通过Fleiss卡方测量的评分者间一致性范围为0.62至0.83。逻辑性和准确性的平均得分分别为24.27±0.4和24.46±0.25。雷达图显示在各项标准上表现始终很高。GPT-4在制定个性化治疗方案、改进手术部位感染管理策略以及识别新出现的研究趋势方面表现出专业能力。
GPT-4在脊柱手术的手术部位感染管理中显示出巨大潜力，促进了以患者为中心的护理和精准医学。尽管在抗生素和患者教育方面存在局限性，但GPT-4的持续学习、对数据隐私的关注以及专业协作表明其有潜力彻底改变手术部位感染管理，需要进一步发展。
Zotero 插件