原发性肝癌放疗后乙肝病毒再激活预测模型的特征降维分析

doi:10.3969/j.issn.0258-8021.2017.06.009

Abstract
Figure/Table
References
Related Citation (3)

Download: PDF (819 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract This study established a classification prediction model for hepatitis B virus (HBV) reactivation after the precise radiotherapy in patients with primary liver cancer (PLC), which is expected to use to prevent HBV reactivation and reduce the incidence of the disease. Ninety of HBV-related HCC patients after receiving precise radiotherapy were recruited from Shandong Cancer Hospital. Each sample tests involved thirty characteristics of sexuality, age, KPS score, AFP level, HBV DNA level, tumor stage TNM etc. In this paper, we proposed a sequential feature selection (SFS) method to select key features which would be combined into a brand new feature subset and then establish Bayesian classification prediction model. The method of sequential backward selection (SBS) showed that the KPS score, HBV DNA, outer margin of radiotherapy, TNM, and total hepatic maximum dose were the risk factors that lead to HBV reactivation. The classification accuracy of Bayesian classification reached to 85.75% using 3 fold cross validation. Besides, Sequential forward selection showed that the sexuality, KPS score, HBV DNA, HBeAg and the two kinds of code of outer margin of radiotherapy were the risk factors that lead to HBV reactivation, meanwhile, the classification accuracy of Bayesian classification reached to 84.06% with 5 fold cross validation. The experimental results showed that the Bayesian classification could be used to study the reactivation of HBV. The key feature had a better classification performance after the feature selection.

Key words： hepatitis B virus(HBV) reactivation sequential feature selection bayesian classification

Received: 23 November 2016

PACS:

TP391

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Wang Huina
	Huang Wei
	Liu Yihui

Cite this article:

Wang Huina,Huang Wei,Liu Yihui. Feature Dimension Reduction of Hepatitis B Virus Reactivation Prediction Model in Patients with Primary Liver Cancer after Precise Radiotherapy[J]. Chinese Journal of Biomedical Engineering, 2017, 36(6): 697-701.

URL:

http://cjbme.csbme.org/EN/10.3969/j.issn.0258-8021.2017.06.009 OR http://cjbme.csbme.org/EN/Y2017/V36/I6/697

[1] 黄伟, 卢彦达, 张炜, 等. 原发性肝癌精确放疗致乙型肝炎病毒再激活分析 [J]. 中华放射肿瘤学杂志, 2013, 22(3):193-197.
[2] Huang Wei, Zhang Wei, Fan Min, et al. Risk factors for hepatitis B virus reactivation after conformal radiotherapy in patients with hepatocellular carcinoma [J]. Cancer Science, 2014, 105(6): 697-703.
[3] 汪孟森. 原发性肝癌三维适形放疗致乙型肝炎病毒再激活相关研究 [D]. 济南: 济南大学, 2014.
[4] 张晶晶,曲颂,余建荣,等. 原发性肝癌三维适形放疗致乙型肝炎病毒再激活相关研究 [J]. 癌症进展,2015(2):183-187.
[5] 吴冠朋,王帅,黄伟,等. 基于BP神经网络的肝癌放疗致乙型肝炎病毒再激活分类预测模型 [J]. 智能计算机与应用,2016, 6(2):43-47.
[6] Wang Shuai, Wu Guanpeng, Huang Wei, et al. The predictive model of hepatitis B virus reactivation induced by precise radiotherapy in primary liver cancer [J]. Journal of Electrical and Electronic Engineering, 2016, 4(2): 31-34.
[7] Wu Guanpeng, Wang Shuai, Huang Wei, et al. Application of BP and RBF neural network in classification prognosis of hepatitis B virus reactivation [J]. Journal of Electrical and Electronic Engineering, 2016, 4(2): 35-39.
[8] Wu Guanpeng, Liu Yihui, Wang Shuai, et al. The classification prognosis models of hepatitis b virus reactivation based on Bayes and support vector machine after feature extraction of genetic algorithm[C]//International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery. Changsha: IEEE, 2016:572-577.
[9] 吴冠朋,刘毅慧,王帅,等. 基于遗传算法特征选择的 HBV 再激活分类预测模型 [J]. 生物信息学,2016,14(4):243-248.
[10] 游伟. 基于支持向量机的基因选择算法研究 [D]. 长沙: 湖南大学,2010.
[11] 游伟,李树涛,谭明奎. 基于SVM-RFE-SFS的基因选择方法[J]. 中国生物医学工程学报, 2010, 29(1):93-99.
[12] Levner I. Feature selection and nearest centroid classification for protein mass spectrometry [J]. BMC Bioinformatics, 2005, 6(1):1-14.
[13] 王姝勤. 肝脏CT辅助诊断系统中特征选择和提取研究 [D]. 上海: 上海交通大学,2010.
[14] Tomar D, Agarwal S. Hybrid feature selection based weighted least squares twin support vector machine approach for diagnosing breast cancer, hepatitis, and diabetes [J]. Advances in Artificial Neural Systems, 2015, 2015:1-10.
[15] 钟珞,潘昊,封筠,等. 模式识别 [M]. 武汉: 武汉大学出版社,2006:138.
[16] 杨淑莹,张桦. 模式识别与智能计算: Matlab技术实现 [M].北京:电子工业出版社,2015.4:67-73.
[17] 王静. 基于贝叶斯的人脸识别 [D]. 郑州: 郑州大学,2006.
[18] 刘丹,方卫国,周泓. 基于贝叶斯网络的二元语法中文分词模型 [J]. 计算机工程,2010,36(1):12-14.
[19] 姚晖,龚金兰,李莉,等. 肝癌患者精确放疗后 HBV 病毒再激活的危险因素分析 [J]. 实用癌症杂志,2014(6):675-677.