目的: 探讨基于多层螺旋CT（multisliece helical CT,MSCT）的计算机纹理分析联合血清神经元特异性烯醇化酶（neuron-specific enolase,NSE）、细胞角蛋白19片段抗原21-1（cytokeratin 19 fragment antigen 21-1,CYFRA21-1）、癌胚抗原（carcinoembryonic antigen,CEA）对孤立性肺结节（solitary pulmonary nodules,SPN）良恶性的鉴别诊断价值。方法: 选取我院2023年1月至2024年12月收治经CT检查疑似恶性风险的126例孤立性肺结节患者,按病理诊断结果分为肺癌组和良性结节组。其中肺癌组65例,肺部良性结节61例。比较两组的一般资料及影像学特征,基于MSCT的计算机纹理参数及血清NSE、CYFRA21-1、CEA的差异,分析肺癌患者计算机纹理特征与血清NSE、CYFRA21-1、CEA的相关性,利用受试者工作特征曲线（receiver operating characteristic,ROC）探讨计算机纹理特征与血清肿瘤标志物各指标单独或联合应用对SPN良恶性的鉴别诊断价值。结果: 肺癌组在结节类型及结节边缘形态方面部分实性结节、分叶毛刺占比高于良性组;肺癌组的熵、熵差、熵和及血清NSE、CYFRA21-1、CEA均高于良性组;且肺癌组计算机纹理相关特征与血清NSE、CYFRA21-1及CEA均呈正相关;ROC曲线结果显示,计算机纹理特征、NSE、CYFRA21-1、CEA及联合应用的AUC分别为0.811、0.710,0.632、0.618及0.951。结论: 计算机纹理特征联合常规血清肿瘤标志物对肺结节良恶性的鉴别诊断力较高,具有较好的临床应用价值。
Abstract
Objective To investigate the diagnostic value of multisliece helical CT (MSCT) based computer texture analysis combined with serum neuron-specific enolase (NSE), cytokeratin 19 fragment antigen 21-1 (CYFRA21-1), and carcinoembryonic antigen (CEA) for distinguishing between benign and malignant solitary pulmonary nodules. Methods A total of 126 patients with solitary pulmonary nodules admitted to our hospital from January 2023 to December 2024 were selected and divided into lung cancer group (65 cases) and benign nodule group (61 cases) based on pathological diagnosis. The general data and imaging characteristics of both groups were compared. The correlation between computer texture parameters from MSCT and serum NSE, CYFRA21-1, and CEA was analyzed. ROC curves were used to evaluate the diagnostic value of computer texture features and serum tumor markers in distinguishing between benign and malignant solitary pulmonary nodules. Results Lung cancer patients showed higher proportions of partially solid nodules and spiculated margins compared to benign patients. Lung cancer patients exhibited higher entropy, entropy difference, entropy sum, and serum NSE, CYFRA21-1, and CEA levels than benign patients. Computer texture-related features showed positive correlations with serum NSE, CYFRA21-1, and CEA in lung cancer patients. ROC curve analysis revealed AUC values of 0.811, 0.710, 0.632, 0.618, and 0.951 for computer texture features, NSE, CYFRA21-1, CEA, and their combined applications respectively. Conclusion The combination of computer texture features and conventional serum tumor markers has a high power of differential diagnosis between benign and malignant pulmonary nodules, and has a good clinical application value.
关键词
基于MSCT的计算机纹理 /
孤立性肺结节 /
鉴别诊断
Key words
MSCT-based computerised texture /
isolated pulmonary nodules /
differential diagnostic
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] NEAD M A, DONY C, RIVERA MP.Diagnosis of the solitary pulmonary nodule[J]. Clin Chest Med, 2025, 46(2): 271-288.
[2] YANG Y C, WU CJ, HSIEH MI, et al.Solitary pulmonary nodule in a renal transplant recipient[J]. J Microbiol Immunol Infect, 2023, 56(4): 883-885.
[3] OSHIMA K.Clinical characteristics of human pulmonary dirofilariasis in japan: an uncommon differential diagnosis of a solitary pulmonary nodule[J]. Jpn J Infect Dis, 2023, 76(5): 310-313.
[4] 黄玉伟, 牛云丽, 王东, 等. 计算机纹理分析技术在鉴别孤立性肺结节多层螺旋CT图像中的临床应用价值[J]. 海南医学, 2023, 34(15): 2213-2216.
[5] 庄献鹏, 赵建峰. CT联合CEA在肺磨玻璃样结节诊断的价值[J]. 中国实验诊断学, 2024, 28(8): 921-923.
[6] 胡泊. AFP、CYFRA21-1、ProGRP水平联合检测在恶性孤立性肺结节诊断中的价值[J]. 中国民康医学, 2023, 35(21): 107-109.
[7] GILLIES RJ, KINAHAN PE, HRICAK H.Radiomics: Images Are More Than Pictures, They Are Data[J]. Radiology, 2016, 278(2): 563-577.
[8] JOKERST C, ADLER C, GOTWAY M, et al.Dual energy technique adds value to solitary pulmonary nodule analysis with dynamic contrast-enhanced ct: a 100 nodule experience[J]. Curr Problems Diagn Radiol, 2023, 52(1): 25-30.
[9] 赵佳佳, 邢媛媛, 王爱辉, 等. PET/CT超级迭代技术诊断孤立性肺结节[J]. 中国医学影像技术, 2023, 39(8): 1201-1205.
[10] 温笑楠, 孙建男, 阎冬琪. CT纹理分析技术在孤立性肺结节良恶性鉴别诊断中的研究进展[J]. 中国医疗设备, 2024, 39(1): 166-171.
[11] 甘荣坤, 陈思敏, 刘昌华. MSCT灌注成像参数鉴别诊断孤立性肺结节良恶性的价值[J]. 沈阳医学院学报, 2024, 26(1): 72-75.
[12] 陈晓雪, 李可峰, 韩海森, 等. CT能谱成像参数联合血清细胞角蛋白19片段抗原21-1水平对孤立性肺结节性质的鉴别诊断价值[J]. 实用心脑肺血管病杂志, 2024, 32(9): 95-98.
[13] 胡翔宇, 沈天赐, 王洋洋, 等. 基于机器学习探究临床联合增强CT影像组学特征对肺结节良恶性的鉴别诊断价值[J]. 湖北医药学院学报, 2024, 43(1): 39-45.
[14] KORKMAZ ET, KOKSAL D, AKSU F, et al.Triple test with tumor markers CYFRA 21.1, HE4, and ProGRP might contribute to diagnosis and subtyping of lung cancer[J]. Clin Biochem, 2018, 58: 15-19.
[15] 王运, 朱红洲, 高心逸, 等. PET/CT中18F-FDG摄取值联合血清CEA水平预测肺腺癌EGFR突变的价值[J]. 浙江医学, 2023, 45(17): 1830-1835.
[16] 王敏, 李丽, 张极平. 瘤内及瘤周联合影像组学模型对肺纯磨玻璃结节浸润性的预测价值[J]. 医学影像学杂志, 2023, 33(6): 975-978.
基金
湖南省自然科学基金医卫行业联合基金项目“嗅鞘细胞调控自噬促进放射性脊髓炎修复的机制研究”（2024JJ9488）
PDF(1634 KB)
