【摘要】 目的 探讨超声联合人工智能技术评估肌少症的诊断价值。方法 选取2020年10月到2022年10月我院收治的30例肌少症患者进行回顾性分析，作为肌少症组，另选取同期来我院体检的30例非肌少症志愿者作为非肌少症组。对比两组患者临床资料，所有患者均采取超声联合人工智能技术进行评估，对比两组患者股四头肌厚度，以CT检测结果，计算骨骼肌指数(SMI)作为金标准，采用Spearman分析肌肉厚度与SMI的相关性。最后，建立ROC曲线与金标准进行比较，分析超声联合人工智能技术评估对肌少症的诊断效能。结果 两组患者性别、年龄、体质量指数、合并高血压、合并糖尿病、合并心衰史、吸烟史、饮酒史及空腹血糖、总胆固醇、高密度脂蛋白胆因醇、低密度脂蛋白胆固醇、前白蛋白、血肌酐、心肌型肌酸激酶同工酶表达水平等项目对比无明显差异(P>0.05);肌少症组股四头肌骨骼肌厚度、SMI明显低于非肌少症组(P<0.05);Spearman相关分析结果显示：股四头肌骨骼肌厚度与SMI水平呈正相关(r=0.579,P<0.018);超声联合人工智能技术中股四头肌骨骼肌厚度对肌少症诊断的最佳阈值为11.23mm。超声联合人工智能技术评估对肌少症的诊断灵敏度与特异度分别为88.26%与95.63%,且与CT金标准诊断对比无统计学差异(P>0.05)。结论 超声联合人工智能技术对肌少症的诊断灵敏度和特异度较高，且有希望成为肌少症诊断的简单、有效方式，为患者的临床干预提供可靠依据。更多还原

【Abstract】 Objective To explore the diagnostic value of ultrasound combined with artificial intelligence technology in the diagnosis of sarcopenia.Methods A retrospective analysis was conducted on 30 patients with sarcopenia admitted to our hospital from October 2020 to October 2022,who were divided into the sarcopenia group.In addition, 30 non sarcopenia volunteers who came to our hospital for physical examination during the same period were selected as the non sarcopenia group.Compare the clinical data of two groups of patients, and evaluate all patients using ultrasound combined with artificial intelligence technology.Compare the thickness of the quadriceps femoris muscle between the two groups of patients, and use CT detection to calculate the skeletal muscle index(SMI)as the gold standard.Use Spearman correlation analysis to analyze the correlation between muscle thickness and SMI.Finally, establish a ROC curve and compare it with the gold standard to analyze the diagnostic efficacy of ultrasound combined with artificial intelligence technology for sarcopenia.Results There was no significant difference between the two groups in gender, age, BMI,hypertension, diabetes, heart failure, smoking, drinking, fasting blood glucose, TC, HDL,LDL,prealbumin, serum creatinine, and CK-MB expression levels(P>0.05).The skeletal muscle thickness and SMI of the quadriceps femoris in the sarcopenia group were significantly lower than those in the non sarcopenia group(P<0.05).The Spearman correlation analysis results showed that the thickness of quadriceps femoris skeletal muscle was positively correlated with SMI level(r=0.579,P<0.018).The optimal threshold for diagnosing sarcopenia using ultrasound combined with artificial intelligence technology is 11.23mm.The diagnostic sensitivity and specificity of ultrasound combined with artificial intelligence technology for sarcopenia were 88.26% and 95.63%,and there was no difference compared to the CT gold standard diagnosis(P>0.05).Conclusion Ultrasound combined with artificial intelligence technology has high diagnostic sensitivity and specificity for sarcopenia, and has the potential to become a simple and effective method for diagnosing sarcopenia, providing reliable basis for clinical intervention for patients.更多还原