MRI纹理分析与LI-RADS分类在肝硬化小肝癌与增生结节诊断中的应用

Application of MRI texture analysis and LI-RADS classification in the diagnosis of small hepatocellular carcinoma and hyperplastic nodules in cirrhosis

目的:探究磁共振检查(MRI)纹理分析与肝脏影像报告和数据系统(liver imaging report and data system,LI-RADS)分类在肝硬化小肝癌与增生结节诊断中的应用价值。方法:回顾性分析烟台芝罘医院2021年1月—2023年1月接诊的60例肝硬化患者(肝硬化结节80个)的相关资料,所有患者均进行MRI检测,依据2018版LI-RADS分类标准对其进行分类分析并分级。使用Logistic回归分析选出最为关键的纹理特征,并构建与之对应的影像组学标签。采用受试者操作特征(receiver operating characteristic,ROC)曲线评估LI-RADS分类结果以及影像组学标签对两种疾病的诊断结果。结果:影像组学标签检出肝硬化结节75个,其中小肝癌59个,增生结节16个;LIRADS分类检出肝硬化结节74个,其中小肝癌59个,增生结节15个。ROC曲线分析显示:影像组学标签鉴定小肝癌和增生结节的灵敏度为95.16%,特异度88.89%,准确率93.75%,Kappa=0.91;LI-RADS分类鉴定小肝癌和增生结节的灵敏度为95.16%,特异度83.33%,准确率92.50%,Kappa=0.95,两种方法诊断灵敏度、特异度、准确率比较差异无统计学意义(P>0.05)。结论:将MRI纹理分析和LI-RADS分类应用在肝硬化小肝癌与增生结节诊断中,对病情的诊断结果更加精准,准确率会更高,可以有效地提高对此类疾病的鉴别和治疗水平,值得在临床医学中推广和使用。

Objective To explore the application value of magnetic resonance imaging(MRI)texture analysis and liver imaging report and data system(LI-RADS)classification in the diagnosis of small hepatocellular carcinoma and hyperplastic nodules in cirrhosis.Methods The relevant data of 60 patients with cirrhosis(80 cirrhosis nodules)admitted to Yantai Zhifu Hospital from January 2021 to January 2023 were retrospectively analyzed.MRI detection was performed on all patients,and classification analysis and grading were performed according to the 2018 edition of LI-RADS classification standard.Logistic regression analysis was used to select the most critical texture features,and the corresponding image omics labels were constructed.Receiver operating characteristic(ROC)curve was used to evaluate the LI-RADS classification results and the diagnosis results of the two diseases with the imaging omics label.Results 75 nodules of liver cirrhosis were detected by imaging group omics label,including 59 small hepatocellular carcinoma and 16 hyperplastic nodules;LI-RADS classification detected 74 cirrhotic nodules,including 59 small hepatocellular carcinoma and 15 hyperplastic nodules.ROC curve analysis showed that the sensitivity,specificity and accuracy of imaging omics labeling for small liver cancer and proliferative nodules were 95.16%,88.89%,93.75%,and Kappa=0.91;The sensitivity,specificity and accuracy of LI-RADS classification were 95.16%,83.33%,92.50%,and Kappa=0.95,there was no statistically significant difference in the sensitivity,specificity and accuracy of the two methods(P>0.05).Conclusion The application of MRI texture analysis and LI-RADS classification in the diagnosis of small hepatocellular carcinoma and hyperplastic nodules in liver cirrhosis can make the diagnosis more accurate and accurate,and can effectively improve the level of differentiation and treatment of such diseases,which is worthy of promotion and use in clinical medicine.