论文部分内容阅读
目的:用人工神经网络(ANN)建立肺癌的4种血清肿瘤标志物(TM)诊断模型,提高诊断肺癌灵敏度的同时保证较高的特异性。方法:检测57例肺癌患者、30例肺部良性疾病患者和29例健康体检者血清标本中癌胚抗原(CEA)、糖类抗原125(CA125)、糖类抗原199(CA199)、鳞状细胞癌相关抗原(SCC)的含量,结合ANN进行数据建模分析。结果:肺癌组与正常组、肺良性疾病组的CEA水平差别有统计学意义(P<0.05),肺癌组与正常组的CA125水平差别有统计学意义(P<0.05),而肺良性疾病组与正常组CA199,SCC水平与肺癌组无差别无统计学意义(P>0.05)。结论:ANN结合4种TM建立的诊断模型分析法提高了肺癌的灵敏度,并保证了较高的特异性,提高了4种TM的临床诊断价值。
OBJECTIVE: To establish a diagnostic model for four kinds of serum tumor markers (TM) of lung cancer by using artificial neural network (ANN) to improve the sensitivity of lung cancer diagnosis and ensure high specificity. Methods: The serum CEA, CA125, CA199, squamous cell count (ESR) and tumor necrosis factor (CDK) of 57 patients with lung cancer, 30 patients with benign pulmonary disease and 29 healthy controls were detected by ELISA. Cancer-associated antigen (SCC) content, combined with ANN for data modeling analysis. Results: The levels of CEA in lung cancer group were significantly different from those in normal group and benign lung disease group (P <0.05), while the levels of CA125 in lung cancer group and normal group were significantly different (P <0.05) There was no significant difference between CA199, SCC and normal lung cancer (P> 0.05). Conclusion: The diagnostic model established by ANN combined with four kinds of TM can improve the sensitivity of lung cancer, ensure high specificity and improve the diagnostic value of four kinds of TM.