论文部分内容阅读
目的观察胃肠肿瘤标志物联合人工神经网络对大肠癌的预警效果。方法将4项胃肠肿瘤标志物进行联合检测,应用人工神经网络技术建立大肠癌肿瘤标志物预警模型(即大肠癌-大肠息肉-正常人的人工神经网络模型),应用该模型和4项胃肠肿瘤标志物联合检测,分别对大肠癌患者进行预测并构建ROC曲线。结果大肠癌-大肠息肉-正常人的人工神经网络模型对大肠癌预测的灵敏度为80.03%,特异度为87.01%,准确度为81.77%,优于肿瘤标志物的联合检测,两者的ROC曲线下面积相比较,差异有统计学意义(P<0.05)。结论 4项胃肠肿瘤标志物检测联合人工神经网络模型,提高了对大肠癌的预测准确性,解决了大量复杂、繁琐的数据分析工作,其操作简便,易于推广和应用。
Objective To observe the early warning effect of gastrointestinal tumor markers combined with artificial neural network on colorectal cancer. Methods Four gastrointestinal tumor markers were detected. Artificial neural network technique was used to establish the early warning model of colorectal cancer tumor markers (ie, artificial neural network model of colorectal cancer - colorectal polyps - normal subjects). Using this model and four stomach Intestinal tumor markers combined detection, respectively, in patients with colorectal cancer prediction and construction of ROC curve. Results The sensitivity of colorectal cancer - colorectal polyps - normal artificial neural network model to prediction of colorectal cancer was 80.03%, specificity was 87.01%, accuracy was 81.77%, better than the combined detection of tumor markers, the ROC curve Under the area compared, the difference was statistically significant (P <0.05). Conclusions Four gastrointestinal tumor markers combined with the artificial neural network model improve the accuracy of the prediction of colorectal cancer and solve a large number of complicated and cumbersome data analysis work, which is easy to operate and easy to popularize and apply.