论文部分内容阅读
目的检测肺癌患者血清中的肺癌肿瘤标记物,应用人工神经网络建立肺癌血清肿瘤标记物诊断模型。方法应用酶联免疫吸附法(ELISA)分别测定86例肺癌和80例健康人血清细胞角蛋白21-1片段(CYFRA21-1)、神经元特异性烯醇化酶(NSE)、组织多肽特异性抗原(TPS)、可溶性白细胞介素2受体(sIL-2R)、癌胚抗原(CEA)、糖抗原242(CA242)、抑癌基因p53抗体共7种肿瘤标记物含量。用曲线下面积结合人工神经网络建立诊断模型,并将此诊断模型用于肺癌的诊断。结果根据血清肿瘤标记物的测定结果,计算出每个肿瘤标记物的曲线下面积,应用人工神经网络建立肺癌血清肿瘤标记物诊断模型,该模型预测肺癌样本的诊断准确率84.1%,敏感性为86.3%,特异性94.8%。结论本研究建立的肺癌诊断模型对肺癌的诊断具有较高的敏感性和特异性。
Objective To detect lung cancer tumor markers in serum of patients with lung cancer and to establish a diagnostic model of lung cancer serum tumor markers by using artificial neural network. Methods Serum cytokeratin 21-1 fragment (CYFRA21-1), neuron specific enolase (NSE), tissue polypeptide specific antigen (TACE) were detected in 86 lung cancer patients and 80 healthy people by enzyme linked immunosorbent assay (ELISA) (TPS), soluble interleukin 2 receptor (sIL-2R), carcinoembryonic antigen (CEA), carbohydrate antigen 242 (CA242) and tumor suppressor p53 antibody. Using the area under the curve combined with artificial neural network to establish a diagnostic model, and the diagnostic model for the diagnosis of lung cancer. Results According to the results of serum tumor markers, the area under the curve of each tumor marker was calculated. The diagnosis model of serum tumor markers was established by using artificial neural network. The diagnostic accuracy of this model was 84.1%, the sensitivity was 86.3%, specificity 94.8%. Conclusion The lung cancer diagnosis model established in this study has high sensitivity and specificity for the diagnosis of lung cancer.