改进的属性约简算法及其在肝癌微血管侵犯预测中的应用

doi:10.11772/j.issn.1001-9081.2019051108

计算机应用 ›› 2019, Vol. 39 ›› Issue (11): 3221-3226.DOI: 10.11772/j.issn.1001-9081.2019051108

• 2019年中国计算机学会人工智能会议(CCFAI2019)论文 • 上一篇下一篇

改进的属性约简算法及其在肝癌微血管侵犯预测中的应用

谭永奇¹, 樊建聪^1,2, 任延德³, 周晓明³

1. 山东科技大学计算机科学与工程学院, 山东青岛 266590;
2. 山东省智慧矿山信息技术重点实验室, 山东青岛 266590;
3. 青岛大学附属医院, 山东青岛 266555

收稿日期:2019-05-24 修回日期:2019-07-18 出版日期:2019-11-10 发布日期:2019-09-11
通讯作者: 樊建聪
作者简介:谭永奇(1994-),男,山东济南人,硕士研究生,CCF会员,主要研究方向:数据挖掘、机器学习;樊建聪(1977-),男,山东青岛人,教授,博士,CCF会员,主要研究方向:数据挖掘、机器学习;任延德(1973-),男,山东泰安人,副主任医师,博士,主要研究方向:神经影像学;周晓明(1977-),男,山东青岛人,副主任医师,硕士,主要研究方向:腹部影像学诊断。
基金资助:
国家重点研发计划项目（2017YFC0804406）；山东省自然科学基金资助项目（ZR2018MF009）；山东省"泰山学者"攀登计划项目。

Improved attribute reduction algorithm and its application to prediction of microvascular invasion in hepatocellular carcinoma

TAN Yongqi¹, FAN Jiancong^1,2, REN Yande³, ZHOU Xiaoming³

1. College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao Shandong 266590, China;
2. Provincial Key Laboratory for Information Technology of Wisdom Mining of Shandong Province, Qingdao Shandong 266590, China;
3. The Affiliated Hospital of Qingdao University, Qingdao Shandong 266555, China

Received:2019-05-24 Revised:2019-07-18 Online:2019-11-10 Published:2019-09-11
Supported by:
This work is partially supported by the National Key Research and Development Program of China (2017YFC0804406), the Shandong Natural Science Foundation (ZR2018MF009), the "Taishan Scholar" Climbing Plan in Shandong Province.

摘要/Abstract

摘要： 基于邻域粗糙集的属性约简算法在进行属性约简时只考虑单一属性对决策属性的影响，未能考虑各属性间的相关性，针对这个问题，提出了一种基于卡方检验的邻域粗糙集属性约简算法（ChiS-NRS）。首先，利用卡方检验计算相关性，在筛选重要属性时考虑相关属性之间的影响，在降低时间复杂度的同时提高了分类准确率；然后，将改进的算法与梯度提升决策树（GBDT）算法组合以建立分类模型，并在UCI数据集上对模型进行验证；最后，将该模型应用于预测肝癌微血管侵犯的发生。实验结果表明，与未约简、邻域粗糙集约简等几种约简算法相比，改进算法在一些UCI数据集上的分类准确率最高；在肝癌微血管侵犯预测中，与卷积神经网络（CNN）、支持向量机（SVM）、随机森林（RF）等预测模型相比，提出的模型在测试集上的预测准确率达到了88.13%，其灵敏度、特异度和受试者操作曲线（ROC）的曲线下面积（AUC）分别为87.10%、89.29%和0.90，各指标都达到了最好。因此，所提模型能更好地预测肝癌微血管侵犯的发生，能辅助医生进行更精确的诊断。

关键词: 属性约简, 卡方检验, 梯度提升树, 微血管侵犯, 邻域粗糙集

Abstract: Focused on the issue that the attribute reduction algorithm based on neighborhood rough set only considers the influence of a single attribute on the decision attribute, and fails to consider the correlation among different attributes, a Neighborhood Rough Set attribute reduction algorithm based on Chi-square test (ChiS-NRS) was proposed. Firstly, the Chi-square test was used to calculate the correlation, and the influence between the related attributes was considered when selecting the important attributes, making the time complexity reduced and the classification accuracy improved. Then, the improved algorithm and the Gradient Boosting Decision Tree (GBDT) algorithm were combined to establish a classification model and the model was verified on UCI datasets. Finally, the proposed model was applied to predict the occurrence of microvascular invasion in hepatocellular carcinoma. The experimental results show that the proposed algorithm has the highest classification accuracy on some UCI datasets compared with the reduction algorithm without reduction and neighborhood rough set reduction algorithm. In the prediction of microvascular invasion in hepatocellular carcinoma, compared with Convolution Neural Network (CNN), Support Vector Machine (SVM) and Random Forest (RF) prediction models, the proposed model has the prediction accuracy of 88.13% in test set, the sensitivity, specificity and the Area Under Curve (AUC) of Receiver Operating Curve (ROC) of 88.89%, 87.5% and 0.90 respectively are the best. Therefore, the prediction model proposed can better predict the occurrence of microvascular invasion in hepatocellular carcinoma and assist doctors to make more accurate diagnosis.

Key words: attribute reduction, Chi-square test, gradient boosting tree, microvascular invasion, neighborhood rough set

中图分类号:

TP181

谭永奇, 樊建聪, 任延德, 周晓明. 改进的属性约简算法及其在肝癌微血管侵犯预测中的应用[J]. 计算机应用, 2019, 39(11): 3221-3226.

TAN Yongqi, FAN Jiancong, REN Yande, ZHOU Xiaoming. Improved attribute reduction algorithm and its application to prediction of microvascular invasion in hepatocellular carcinoma[J]. Journal of Computer Applications, 2019, 39(11): 3221-3226.

参考文献

[1] FERLAY J, SOERJOMATARAM I, DIKSHIT R, et al. Cancer incidence and mortality worldwide:sources, methods and major patterns in GLOBOCAN 2012[J]. International Journal of Cancer, 2015, 136(5):E359-E386.
[2] BRUIX J, GORES G J, MAZZAFERRO V. Hepatocellular carcinoma:clinical frontiers and perspectives[J]. Gut, 2014, 63(5):844-855.
[3] CHENG Z, YANG P, QU S, et al. Risk factors and management for early and late intrahepatic recurrence of solitary hepatocellular carcinoma after curative resection[J]. HPB, 2015, 17(5):422-427.
[4] ZHAO H, HUA Y, DAI T, et al. Development and validation of a novel predictive scoring model for microvascular invasion in patients with hepatocellular carcinoma[J]. European Journal of Radiology, 2017, 88:32-40.
[5] YANG P, SI A, YANG J, et al. A wide-margin liver resection improves long-term outcomes for patients with HBV-related hepatocellular carcinoma with microvascular invasion[J]. Surgery, 2019, 165(4):721-730.
[6] 马海, 王宇, 杨红春,等. 预测肝癌微血管侵犯及早期复发的临床研究[J]. 中华临床医师杂志(电子版), 2012, 6(20):58-60.(MA H, WANG Y, YANG H C, et al. Clinical study on predicting microvascular invasion and early recurrence of hepatocellular carcinoma[J]. Chinese Journal of Clinical Physicians (Electronic Edition), 2012, 6(20):58-60.)
[7] ZHOU L, RUI J, ZHOU W, et al. Edmondson-Steiner grade:a crucial predictor of recurrence and survival in hepatocellular carcinoma without microvascular invasion[J]. Pathology Research and Practice, 2017, 213(7):824-830.
[8] 刘桐桐, 董怡, 韩红,等. 基于影像组学方法的原发性肝细胞癌微血管侵犯和肿瘤分化等级预测[J]. 中国医学计算机成像杂志, 2018, 24(1):83-87.(LIU T T, DONG Y, HAN H, et al. Prediction of microvascular invasion and tumor differentiation grade in hepatocellular carcinoma based on radiomics[J]. Chinese Journal of Medical Computer Imaging, 2018, 24(1):83-87.)
[9] 武明辉, 谭红娜, 吴青霞,等. 肝脏磁共振T2WI图像纹理特征预测肝细胞癌患者微血管侵犯的价值[J]. 中国癌症杂志, 2018, 28(3):191-196.(WU M H, TAN H N, WU Q X, et al. Value of MRI T2-weighted image texture analysis in evaluating the microvascular invasion for hepatocellular carcinoma[J]. Chinese Journal of Cancer, 2018, 28(3):191-196.)
[10] VELAYUTHAM C, THANGAVEL K. Detection and elimination of pectoral muscle in mammogram images using rough set theory[C]//Proceedings of the 2012 IEEE International Conference on Advances in Engineering, Science and Management. Piscataway:IEEE, 2012:48-54.
[11] XIE Q, ZENG H, RUAN L, et al. Transformer fault diagnosis based on Bayesian network and rough set reduction theory[C]//Proceedings of the 2013 IEEE TENCON Spring Conference. Piscataway:IEEE, 2013:262-266.
[12] 李林霄, 董昱. 基于粗糙集理论和FCM的轨道电路故障诊断模型[J]. 铁道标准设计,2018(12):169-173.(LI L X, DONG Y. Track circuit fault diagnosis model based on principal component heuristic algorithm[J]. Railway Standard Design, 2018(12):169-173.)
[13] WANG S, LI X, ZHANG S, et al. Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction[J]. Computers in Biology and Medicine, 2010, 40(2):179-189.
[14] LIU Y, XIE H, CHEN Y, et al. Neighborhood mutual information and its application on hyperspectral band selection for classification[J]. Chemometrics and Intelligent Laboratory Systems, 2016, 157:140-151.
[15] LI N, ZHOU R, HU Q, et al. Mechanical fault diagnosis based on redundant second generation wavelet packet transform, neighborhood rough set and support vector machine[J]. Mechanical Systems and Signal Processing, 2012,28:608-621.
[16] 徐久成, 徐天贺, 孙林,等. 基于邻域粗糙集和粒子群优化的肿瘤分类特征基因选取[J]. 小型微型计算机系统, 2014, 35(11):2528-2532.(XU J C, XU T H, SUN L, et al. Feature selection for cancer classification based on neighborhood rough set and particle swarm optimization[J]. Journal of Chinese Computer Systems, 2014, 35(11):2528-2532.)
[17] CHEN Y, ZHANG Z, ZHENG J, et al. Gene selection for tumor classification using neighborhood rough sets and entropy measures[J]. Journal of Biomedical Informatics, 2017, 67:59-68.
[18] 胡玮. 基于改进邻域粗糙集和随机森林算法的糖尿病预测研究[D].北京:首都经济贸易大学, 2018:14-16.(HU W. Research on prediction of diabetes based on improved neighborhood rough set and random forest algorithm[D]. Beijing:Capital University of Business and Economics, 2018:14-16.)

改进的属性约简算法及其在肝癌微血管侵犯预测中的应用

Improved attribute reduction algorithm and its application to prediction of microvascular invasion in hepatocellular carcinoma

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王磊. 改进粗糙集属性约简结合K-means聚类的网络入侵检测方法[J]. 计算机应用, 2020, 40(7): 1996-2002.
[2]	章夏杰, 朱敬华, 陈杨. Spark下的分布式粗糙集属性约简算法[J]. 计算机应用, 2020, 40(2): 518-523.
[3]	欧彬利, 钟夏汝, 代建华, 杨田. 基于变精度覆盖粗糙集的入侵检测方法[J]. 计算机应用, 2020, 40(12): 3465-3470.
[4]	李孜颖, 石振国. 面向大数据任务的调度方法[J]. 计算机应用, 2020, 40(10): 2923-2928.
[5]	鲍迪, 张楠, 童向荣, 岳晓冬. 区间值决策表的正域增量式属性约简算法[J]. 计算机应用, 2019, 39(8): 2288-2296.
[6]	陈曼如, 张楠, 童向荣, 东野升龙, 杨文静. 基于多尺度属性粒策略的快速正域约简算法[J]. 计算机应用, 2019, 39(12): 3426-3433.
[7]	李旭, 荣梓景, 阮晓曦. 关系决策系统中相对不可区分和区分关系的约简[J]. 计算机应用, 2019, 39(10): 2852-2858.
[8]	袁钟, 冯山. 基于邻域值差异度量的离群点检测算法[J]. 计算机应用, 2018, 38(7): 1905-1909.
[9]	王映龙, 曾淇, 钱文彬, 舒文豪, 黄锦涛. 变精度下不完备混合数据的增量式属性约简方法[J]. 计算机应用, 2018, 38(10): 2764-2771.
[10]	姚晟, 汪杰, 徐风, 陈菊. 不完备邻域粗糙集的不确定性度量和属性约简[J]. 计算机应用, 2018, 38(1): 97-103.
[11]	郑忠仁, 程勇, 王军, 钟水明, 徐利亚. 基于遗传算法的气象观测数据区间值属性约简算法[J]. 计算机应用, 2017, 37(9): 2678-2683.
[12]	江洋, 李成海. 基于改进变精度粗糙集的漏洞威胁评估[J]. 计算机应用, 2017, 37(5): 1353-1356.
[13]	孙丽华, 严军峰, 徐健锋. 基于多机器学习竞争策略的短时雷电预报[J]. 计算机应用, 2016, 36(9): 2555-2559.
[14]	熊方, 张贤勇. 变精度粗糙集的区域属性约简及其结构启发算法[J]. 计算机应用, 2016, 36(11): 2954-2957.
[15]	罗豪, 续欣莹, 谢珺, 张扩, 谢新林. 基于扩展容差关系的不完备信息系统属性约简[J]. 计算机应用, 2016, 36(11): 2958-2962.