基于数据挖掘的低增生性骨髓增生异常综合征与再生障碍性贫血分类模型研究  

Classification model of hypocellular myelodysplastic syndrome and aplastic anemia based on data mining

作　　者：刘一航 宋洁 李梦洁 陈松 姜勃宇 俎毓伟 张春英 武建辉 LIU Yi-hang;SONG Jie;LI Meng-jie;CHEN Song;JIANG Bo-yu;ZU Yu-wei;ZHANG Chun-ying;WU Jian-hui(School of Public Health,North China University of Technology,Tangshan,Hebei 063210,China)

机构地区：[1]华北理工大学公共卫生学院,河北唐山063210

出　　处：《现代预防医学》2021年第17期3254-3258,共5页Modern Preventive Medicine

基　　金：河北省自然科学基金项目(H2017209172)。

摘　　要：目的构建低增生性骨髓增生异常综合征(hypo-MDS)与再生障碍性贫血(AA)鉴别诊断的决策树、贝叶斯、卷积神经网络、改进的支持向量机四种模型并选择出最优模型。方法收集2010—2019年华北理工大学附属医院的AA与hypo-MDS患者的病例资料,使用统计学方法筛选指标,将处理后的样本以4∶1随机分为训练集和测试集,构建决策树、贝叶斯、卷积神经网络、改进的支持向量机四种模型,采用五折交叉验证法多次重复验证,通过灵敏度、AUC等指标评价鉴别诊断效果。结果 hypo-MDS患者红细胞、血红蛋白含量等指标低于AA患者,成熟单核细胞比例等指标高于AA患者,年龄和职业分布也存在差异(P<0.05);最终选出21个特异性指标。四种模型的分类效果比较:灵敏度分别为82.56%、65.12%、87.21%、79.07%;AUC分别为0.81、0.68、0.82、0.83;准确率分别为75.32%、69.48%、77.27%、74.03%。对卷积神经网络的误判病例分析得出年龄、血成熟淋巴细胞等7个指标均存在差异(P<0.05)。结论在决策树、贝叶斯、卷积神经网络、改进的支持向量机四种诊断模型中,卷积神经网络具有最佳分类效果。Objective To construct four models of Decision Tree,Bayes,Convolutional Neural Network,Genetic Algorithm of the Support Vector Machines for the differential diagnosis of hypocellular myelodysplastic syndrome(hypo-MDS)and aplastic anemia(AA)and choose the best model.Methods The case data of AA and hypo-MDS patients of the Affiliated Hospital of North China University of Science and Technology from 2010 to 2019 was collected,statistical methods to screen indicators were used,and the processed samples were divided into training set and test set randomly at 4∶1 to construct decision trees and shellfish.The four models of Decision Tree,Bayes,Convolutional Neural Network,Genetic Algorithm of the Support Vector Machines repeated the verification many times by using the five-fold cross-validation method,and we evaluated the differential diagnosis effect through indicators such as sensitivity and AUC.Results The red blood cell and hemoglobin content of hypo-MDS patients were lower than those of AA patients,and the ratio of mature monocytes was higher than that of AA patients.There were also differences in age and occupational distribution(P<0.05).21 specific indicators were finally selected.Comparison of the classification effects of the four models were as follows:sensitivities were 82.56%,65.12%,87.21%,79.07%,AUC were 0.81,0.68,0.82,0.83 and accuracy rates were 75.32%,69.48%,77.27%,74.03%.Analysis of the misjudgment cases of the convolutional neural network showed that there were differences in 7 indicators including age and blood mature lymphocytes(P<0.05).Conclusion Among the four diagnostic models of Decision Tree,Bayes,Convolutional Neural Network,Genetic Algorithm of the Support Vector Machines,Convolutional Neural Network has the best classification effect.

关 键 词：骨髓增生异常综合征 再生障碍性贫血 决策树 贝叶斯 卷积神经网络 改进的支持向量机 

分 类 号：R551.3[医药卫生—血液循环系统疾病] R556.5[医药卫生—内科学]