论文部分内容阅读
目的:探讨模式识别及人工神经网络技术在肺癌组织分型中的应用。方法:用放射性免疫法测定了肺癌患者血清中4种肿瘤标志物(CEA、CA125、胃泌素及NSE)的水平,在此基础上采用模式识别及人工神经网络技术,探讨它们在肺癌组织分型中的应用价值。结果:在判别小细胞肺癌与非小细胞肺癌类型中,这些方法的总正确率均在85％以上。结论:模式识别及人工神经网络技术在肺癌组织分型中有一定的参考价值,同时为临床提供必要的参考资料。
Objective: To explore the application of pattern recognition and artificial neural network in lung cancer tissue typing. Methods: The levels of four tumor markers (CEA, CA125, gastrin and NSE) in sera of patients with lung cancer were determined by radioimmunoassay. Based on this, pattern recognition and artificial neural network were used to detect the levels of four tumor markers Type of application value. Results: The overall accuracy of these methods was over 85% in distinguishing small cell lung cancer from non-small cell lung cancer. Conclusion: Pattern recognition and artificial neural network technology have certain reference value in lung cancer tissue typing, and provide the necessary reference for clinical application.