论文部分内容阅读
目的:应用人工神经网络技术,联合检测6种肿瘤标志对肺癌与胃癌或肠癌进行区分判别,建立肿瘤标志联合检测肺癌的辅助诊断模型。方法:采用放射免疫学、分光光度法、原子吸收分光光度法等方法,测定67例肺癌患者、47例胃癌患者和50例大肠癌患者血清中癌胚抗原(CEA)、胃泌素(gastrin)、神经元特异性烯醇化酶(NSE)、唾液酸(SA)、铜锌比值(Cu/Zn)、钙(Ca)等6项指标。建立基于人工神经网络的肺癌肿瘤标志智能诊断模型。结果:肺癌-胃癌的人工神经网络模型判别肺癌的灵敏度,特异度和准确度分别为100%、83.3%和93.5%;肺癌-肠癌模型判别肺癌的灵敏度、特异度和准确度分别为76.9%、100%和87.0%。结论:本研究成功建立基于人工神经网络技术的肿瘤标志物联合检测的人工智能诊断模型,对肺癌-胃癌、肺癌-肠癌中肺癌的鉴别诊断有助于提高肺癌的诊断率。
OBJECTIVE: To make a differential diagnosis between lung cancer and gastric cancer or intestinal cancer by using six kinds of tumor markers combined with artificial neural network technology, and to establish an auxiliary diagnosis model of tumor marker combined with lung cancer detection. Methods: Serum carcinoembryonic antigen (CEA), gastrin (CD44), CD44 (superscript +) and CD8 + were detected by radioimmunoassay, spectrophotometry and atomic absorption spectrophotometry in 67 patients with lung cancer, 47 patients with gastric cancer and 50 patients with colorectal cancer , Neuron specific enolase (NSE), sialic acid (SA), copper / zinc ratio (Ca / Zn), calcium (Ca) Establishment of an Intelligent Diagnosis Model of Lung Cancer Tumor Markers Based on Artificial Neural Network. Results: The sensitivity, specificity and accuracy of lung cancer-gastric cancer model were 100%, 83.3% and 93.5% respectively. The sensitivity, specificity and accuracy of lung cancer-colon cancer model were 76.9% , 100% and 87.0%. Conclusion: In this study, an artificial intelligence diagnosis model based on artificial neural network (ANN) combined detection of tumor markers was successfully established. The differential diagnosis of lung cancer in lung cancer - gastric cancer, lung cancer - intestinal cancer helps to improve the diagnostic rate of lung cancer.