Research on Feature Extraction Technology of Electronic Medical Record Data Based on Neural Networks

doi:10.3969/j.issn.1671-1122.2023.10.010

Abstract

Abstract:

With the implementation of laws and regulations such as “Data Security Law of the People’s Republic of China”, data security is becoming increasingly important. Electronic medical records contain sensitive personal information such as citizens’ medical and health care. In order to protect the safety of the data, this paper studied the feature extraction technology of the data to provide technical support for the implementation of data security protection. This paper proposed a feature extraction method for electronic medical record data based on deep neural networks. Using generative adversarial networks, a small amount of electronic medical record data was expanded to a larger dataset through text generation methods. Then, the convolutional neural networks were used for feature extraction, and the classification results were generated by the classifier to detect and recognize the electronic medical record data. The experimental results show that this method has a good feature extraction effect for electronic medical record data.

Key words: generative adversarial networks, convolutional neural networks, feature extraction, text generation

CLC Number:

TP309

QIN Yifang, ZHANG Jian, LIANG Chen. Research on Feature Extraction Technology of Electronic Medical Record Data Based on Neural Networks[J]. Netinfo Security, 2023, 23(10): 70-76.

Figures/Tables 7

References 15

[1]	PENG Baolin, ZHU Chenguang, LI Chunyuan, et al. Few-Shot Natural Language Generation for Task-Oriented Dialog[C]// ACL. Findings of the Association for Computational Linguistics:EMNLP. Stroudsburg: ACL, 2020: 172-182.
[2]	IAN J G, JEAN P A, MEHDI M, et al. Generative Adversarial Networks[C]// ACM. NIPS:Neural Information Processing Systems Conference. New York: ACM, 2014: 2672-2680.
[3]	LIU Wenfu. Research on the Key Techniques of Short Text Representation and Classification Based on Mixed Semantic Learning[D]. Zhengzhou: PLA Information Engineering University, 2022.
	刘文甫. 基于混合语义学习的短文本表示与分类关键技术研究[D]. 郑州: 解放军信息工程大学, 2022.
[4]	TOMAS M, KAI C, GREG C, et al. Efficient Estimation of Word Representations in Vector Space[EB/OL]. (2013-09-07) [2023-06-01]. https://arxiv.org/abs/1301.3781.
[5]	ZHANG Li, ZOU Kaihong, ZONG Xu, et al. Text Feature Extraction Based on Deep Convolutional Neural Networks: China, CN201810379548. X[P]. 2018-09-28.
	张黎, 邹开红, 宗旭, 等. 基于深度卷积神经网络进行文本特征提取的方法:中国,CN201810379548. X[P]. 2018-09-28.
[6]	ZHAO Qinlu, CAI Xiaodong, LI Bo, et al. Text Feature Extraction Method Based on LSTM-Attention Neural Networks[J]. Modern Electronic Technology, 2018, 41(8): 167-170.
	赵勤鲁, 蔡晓东, 李波, 等. 基于LSTM-Attention神经网络的文本特征提取方法[J]. 现代电子技术, 2018, 41(8): 167-170.
[7]	REITER E, DALE R, FENG Z. Building Natural Language Generation Systems[M]. Cambridge: Cambridge University Press, 2000.
[8]	WAN Xiaojun, FENG Yansong, SUN Weiwei. Research Progress and Trend of Text Automatic Generation[C]// CCF. CCF 2014-2015 China Computer Science and Technology Development Report. Beijing: CCF, 2015: 298-323.
	万小军, 冯岩松, 孙薇薇. 文本自动生成研究进展与趋势[C]// CCF. CCF 2014-2015中国计算机科学技术发展报告会. 北京: CCF, 2015:298-323.
[9]	HU Baotian. Text Representation Based on Deep Neural Networks and Its Application[D]. Harbin: Harbin Institute of Technology, 2016.
	户保田. 基于深度神经网络的文本表示及其应用[D]. 哈尔滨: 哈尔滨工业大学, 2016.
[10]	LI Tong. Research on Natural Language Generation Algorithm Based on Transformer[D]. Xi’ an: Xidian University, 2022.
	李彤. 基于Transformer的自然语言生成算法研究[D]. 西安: 西安电子科技大学, 2022.
[11]	YU Lantao, ZHANG Weinan, WANG Jun, et al. SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient[C]// AAAI. AAAI Conference on Artificial Intelligence. New York: AAAI, 2017: 2852-2858.
[12]	CHEN Xuesong, YANG Yimin. Review on Reinforcement Learning[J]. Application Research of Computers, 2010, 27(8): 2834-2838, 2844.
	陈学松, 杨宜民. 强化学习研究综述[J]. 计算机应用研究, 2010, 27(8): 2834-2838,2844.
[13]	KIM Y. Convolutional Neural Networks for Sentence Classification. Association for Computational Linguistics[C]// ACL. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). Stroudsburg: ACL, 2014: 1746-1751.
[14]	HU Mianning, LI Xin, LI Mingfeng, et al. Research on Multi-Strategy Data Enhancement Technology for Fraud Short Message Identification[J]. Netinfo Security, 2022, 22(10): 121-128.
	胡勉宁, 李欣, 李明锋, 等. 面向诈骗短信息识别的融合多策略数据增强技术研究[J]. 信息网络安全, 2022, 22(10): 121-128.
[15]	GU Zhaojun, LIU Tingting, GAO Bing, et al. Anomaly Detection of Imbalanced Data in Industrial Control System Based on GAN-Cross[J]. Netinfo Security, 2022, 22(8): 81-89.
	顾兆军, 刘婷婷, 高冰, 等. 基于GAN-Cross的工控系统类不平衡数据异常检测[J]. 信息网络安全, 2022, 22(8): 81-89.

数据类别	总数/个	训练集/个	验证集/个	测试集/个
初始电子病历	5	2	1	1
新增xml数据	10	8	1	1
生成电子病历	200	160	20	20
电影	100	80	10	10
电子竞技	100	80	10	10
星座	100	80	10	10
教育	100	80	10	10
国际	100	80	10	10
总计	715	571	72	72

参数	参数含义	参数设置
embedding_dim	词向量维度	64
seq_length	序列长度	600
num_classes	类别数	7
num_filters	卷积核数目	256
kernel_size	卷积核尺寸	5
vocab_size	词汇表大小	5000
hidden_dim	全连接层神经元	128
dropout_keep_prob	dropout保留比例	0.5
learning_rate	学习率	0.001
batch_size	每批训练大小	64
num_epochs	总迭代轮次	20

类别	Precision	Recall	F1值
初始电子病历	1.00	1.00	1.00
新增xml数据	1.00	1.00	1.00
生成电子病历	0.95	1.00	0.98
电影	1.00	0.60	0.75
电子竞技	0.91	1.00	0.95
星座	0.91	1.00	0.95
教育	1.00	0.90	0.95
国际	0.83	1.00	0.91