基于变分自编码器的异常颈动脉早期识别和预测

doi:10.11772/j.issn.1001-9081.2020101695

计算机应用 ›› 2021, Vol. 41 ›› Issue (10): 3082-3088.DOI: 10.11772/j.issn.1001-9081.2020101695

所属专题：前沿与综合应用

基于变分自编码器的异常颈动脉早期识别和预测

黄晓祥^1,2, 胡咏梅¹, 吴丹², 任力杰³

1. 山东大学控制科学与工程学院, 济南 250061;
2. 中国科学院深圳先进技术研究院, 广东深圳 518055;
3. 深圳市第二人民医院神经内科, 广东深圳 518028

收稿日期:2020-11-02 修回日期:2021-01-28 出版日期:2021-10-10 发布日期:2021-02-24
通讯作者: 胡咏梅
作者简介:黄晓祥(1993-),男,湖北襄阳人,硕士研究生,主要研究方向:机器学习、数据挖掘;胡咏梅(1965-),女,山东济南人,教授,博士,主要研究方向:粗糙集、模式识别、数据挖掘、医院信息系统;吴丹(1984-),女,广东深圳人,助理研究员,博士,主要研究方向:数据挖掘、健康大数据分析与心血管疾病预测;任力杰(1972-),男,广东深圳人,主任医师,博士,主要研究方向:心血管疾病。
基金资助:
国家自然科学基金资助项目（81701788）；深圳市科创委应用示范项目（KJYY20180703165202011）。

Early identification and prediction of abnormal carotid arteries based on variational autoencoder

HUANG Xiaoxiang^1,2, HU Yongmei¹, WU Dan², REN Lijie³

1. School of Control Science and Engineering, Shandong University, Jinan Shandong 250061, China;
2. Shenzhen Institutes of Advanced Technology, Chinese Academy of Science, Shenzhen Guangdong 518055, China;
3. Neurology Department, The Second People's Hospital of Shenzhen, Shenzhen Guangdong 518028, China

Received:2020-11-02 Revised:2021-01-28 Online:2021-10-10 Published:2021-02-24
Supported by:
This work is partially supported by the National Natural Science Foundation of China (81701788), the Shenzhen Science and Technology Innovation Commission Application Demonstration Project (KJYY20180703165202011).

摘要/Abstract

摘要： 颈动脉狭窄、颈动脉内中膜厚度增加（CIMT）或颈动脉斑块等可导致脑卒中的发生。为实现脑卒中大规模初步筛查，提出基于医疗数据的改进的变分自编码器（VAE）来识别和预测异常颈动脉。首先，针对医疗数据存在缺失的情况，采用K近邻（KNN）、均值和众数相混合的方法（MKNN）以及改进的VAE对缺失数据进行填补以得到完整的数据集，从而提高数据的应用范围；接着，分析特征属性，并依据重要性对特征进行排序；然后，运用逻辑回归（LR）、支持向量机（SVM）、随机森林（RF）和极限梯度提升树（XGBT）这四种有监督学习方法结合遗传算法（GA）来建立异常颈动脉识别模型；最后，基于改进的VAE建立预测异常颈动脉的半监督模型。相较于基线模型，基于改进的VAE的半监督模型性能提升明显，灵敏度达到0.893 8，特异性达到0.927 2，F1值达到0.910 5，分类准确率达到0.910 5。实验结果表明，所建立的半监督模型可以用来识别异常颈动脉，进而作为一种识别脑卒中高危人群的工具，预防和减少脑卒中的发生。

关键词: 颈动脉, 脑卒中, 变分自编码器, 遗传算法, 半监督模型

Abstract: Carotid artery stenosis, Carotid Intima Media Thickness (CIMT) or carotid artery plaque may lead to stroke. For large-scale preliminary screening of stroke, an improved Variational AutoEncoder (VAE) based on medical data was proposed to predict and identify abnormal carotid arteries. Firstly, for the missing values in medical data, K-Nearest Neighbor (KNN), Mixture of mean, mode and KNN (MKNN) method and improved VAE were respectively used to impute the missed values to obtain the complete dataset, improving the application range of the data. Secondly, the feature attributes were analyzed and the features were ranked in order of importance. Thirdly, four supervised algorithms, Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF) and eXtreme Gradient Boosting Tree (XGBT), were combined with Genetic Algorithm (GA) to build the abnormal carotid artery identification models. Finally, based on the improved VAE, a semi-supervised abnormal carotid artery prediction model was built. Compared to the performance of baseline model, the performance of the semi-supervised model based on the improved VAE improves significantly with sensitivity of 0.893 8, specificity of 0.927 2, F1-measure of 0.910 5 and classification accuracy of 0.910 5. Experimental results show that this semi-supervised model can be used to identify the abnormal carotid arteries and thus serves as a tool to recognize high-risk groups of stroke, preventing and reducing the occurrence of stroke.

Key words: carotid artery, stroke, Variational AutoEncoder (VAE), Genetic Algorithm (GA), semi-supervised model

中图分类号:

TP391.7

黄晓祥, 胡咏梅, 吴丹, 任力杰. 基于变分自编码器的异常颈动脉早期识别和预测[J]. 计算机应用, 2021, 41(10): 3082-3088.

HUANG Xiaoxiang, HU Yongmei, WU Dan, REN Lijie. Early identification and prediction of abnormal carotid arteries based on variational autoencoder[J]. Journal of Computer Applications, 2021, 41(10): 3082-3088.

参考文献

[1] CAMPBELL B C V, DE SILVA D A, MACLEOD M R, et al. Ischaemic stroke[J]. Nature Reviews Disease Primers, 2019, 5:No. 70.
[2] GBD 2016 Lifetime Risk of Stroke Collaborators. Global, regional, and country-specific lifetime risks of stroke, 1990 and 2016[J]. The New England Journal of Medicine, 2018, 379(25):2429-2437.
[3] EKERETE I, NUGENT C, McLAUGHLIN J. An unobtrusive sensing solution for home based post-stroke rehabilitation[C]//Proceedings of the 2018 IEEE International Conference on Bioinformatics and Biomedicine. Piscataway:IEEE, 2019:1478-1479.
[4] LUDWIG M, VON PETZINGER-KRUTHOFF A, VON BUQUOY M, et al. Intima-Media-Dicke der Karotisarterien:Früher Indikator für Arteriosklerose und therapeutischer Endpunkt[J]. Ultraschall in der Medizin, 2003, 24(3):162-174.
[5] RILEY W A. Cardiovascular risk assessment in individual patients from carotid intimal-medial thickness measurements[J]. Current Atherosclerosis Reports, 2004, 6(3):225-231.
[6] ALI Y S, REMBOLD K E, WEAVER B, et al. Prediction of major adverse cardiovascular events by age-normalized carotid intimal medial thickness[J]. Atherosclerosis, 2006, 187(1):186-190.
[7] PRATI P, VANUZZO D, CASAROLI M, et al. Determinants of carotid plaque occurrence[J]. Cerebrovascular Diseases, 2006, 22(5/6):416-422.
[8] EIGENBRODT M L, BURSAC Z, TRACY R E, et al. B-mode ultrasound common carotid artery intima-media thickness and external diameter:cross-sectional and longitudinal associations with carotid atherosclerosis in a large population sample[J]. Cardiovascular Ultrasound, 2008, 6:No. 10.
[9] KUBALE R, ARNING C. Stellenwert der Doppler-sonographischen Verfahren zur Diagnose der Karotisstenosen[J]. Der Radiologe, 2004, 44(10):946-959.
[10] HOLLAND J H. Adaptation in Natural and Artificial Systems:An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence[M]. Cambridge:MIT Press, 1976:529.
[11] 文艺, 董玉斌. 先天性巨结肠术后并发肠炎危险因素逻辑回归分析[J]. 中华实验外科杂志, 2019, 36(1):130.(WEN Y, DONG Y B. Logistic regression analysis of risk factors for postoperative enteritis in Hirschsprung's disease[J]. Chinese Journal of Experimental Surgery, 2019, 36(1):130.)
[12] 徐静妹, 李雷. 基于稀疏表示和支持向量机的人脸识别算法[J]. 计算机技术与发展, 2018, 28(2):59-63.(XU J M, LI L. A face recognition algorithm based on sparse representation and support vector machine[J]. Computer Technology and Development, 2018, 28(2):59-63.)
[13] VIJIYAKUMAR K, LAVANYA B, NIRMALA I, et al. Random forest algorithm for the prediction of diabetes[C]//Proceedings of the 2019 IEEE International Conference on System, Computation, Automation and Networking. Piscataway:IEEE, 2019:1-5.
[14] HU Z X, QIU H, SU Z Q, et al. A stacking ensemble model to predict daily number of hospital admissions for cardiovascular diseases[J]. IEEE Access, 2020, 8:138719-138729.
[15] SAPOSNIK G, COTE R, MAMDANI M, et al. JURaSSiC:accuracy of clinician vs risk score prediction of ischemic stroke outcomes[J]. Neurology, 2013, 81(5):448-455.
[16] NTAIOS G, GIOULEKAS F, PAPAVASILEIOU V, et al. ASTRAL, DRAGON and SEDAN scores predict stroke outcome more accurately than physicians[J]. European Journal of Neurology, 2016, 23(11):1651-1657.
[17] BRINDLE P, EMBERSON J, LAMPE F, et al. Predictive accuracy of the Framingham coronary risk score in British men:prospective cohort study[J]. BMJ, 2003, 327(7426):No. 1267.
[18] YANG X L, LI J X, HU D S, et al. Predicting the 10-year risks of atherosclerotic cardiovascular disease in Chinese population:the China-PAR project (prediction for ASCVD risk in China)[J]. Circulation, 2016, 134(19):1430-1440.
[19] MIR A, DHAGE S N. Diabetes disease prediction using machine learning on big data of healthcare[C]//Proceedings of the 4th International Conference on Computing Communication Control and Automation. Piscataway:IEEE, 2018:1-6.
[20] KOUROU K, EXARCHOS T P, EXARCHOS K P, et al. Machine learning applications in cancer prognosis and prediction[J]. Computational and Structural Biotechnology, 2015, 13:8-17.
[21] KIM K J, TAGKOPOULOS I. Application of machine learning in rheumatic disease research[J]. The Korean Journal of Internal Medicine, 2019, 34(4):708-722.
[22] ISLAM M M, WU C C, POLY T N, et al. Applications of machine learning in fatty live disease prediction[J]. Studies in Health Technology and Informatics, 2018, 247:166-170.
[23] WENG S F, REPS J, KAI J, et al. Can machine-learning improve cardiovascular risk prediction using routine clinical data?[J]. PLoS ONE, 2017, 12(4):No. e0174944.
[24] KINGMA D P, WELLING M. Auto-Encoding variational Bayes[EB/OL]. (2014-05-01)[2020-12-22]. https://arxiv.org/pdf/1312.6114.pdf.
[25] TU C T, CHEN Y F. Facial Image inpainting with variational autoencoder[C]//Proceedings of the 2nd International Conference of Intelligent Robotic and Control Engineering. Piscataway:IEEE, 2019:119-122.
[26] LIN C C, HUNG Y, FERIS R, et al. Video instance segmentation tracking with a modified VAE architecture[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2020:13144-13154.
[27] BI L Z, ZHANG J W, LIAN J L. EEG-based adaptive drivervehicle interface using variational autoencoder and PI-TSVM[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27(10):2025-2033.
[28] NAZÁBAL A, OLMOS P M, GHAHRAMANI Z, et al. Handling incomplete heterogeneous data using VAEs[J]. Pattern Recognition, 2020, 107:No. 107501.
[29] BARZ B, RODNER E, GARCIA Y G, et al. Detecting regions of maximal divergence for spatio-temporal anomaly detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(5):1088-1101.
[30] FANG Y Q, XIAO X, GE J W. Cloud computing task scheduling algorithm based on improved genetic algorithm[C]//Proceedings of the IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference. Piscataway:IEEE, 2019:852-856.
[31] ÇINAROĞLU S, BODUR S. A new hybrid approach based on genetic algorithm for minimum vertex cover[C]//Proceedings of 2018 IEEE International Conference on Innovations in Intelligent Systems and Applications. Piscataway:IEEE, 2018:1-5.
[32] RESHEF D N, RESHEF Y A, FINUCANE H K, et al. Detecting novel associations in large data sets[J]. Science, 2011, 334(6062):1518-1524.

基于变分自编码器的异常颈动脉早期识别和预测

Early identification and prediction of abnormal carotid arteries based on variational autoencoder

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]	张闻强, 邢征, 杨卫东. 基于多区域采样策略的混合粒子群优化求解多目标柔性作业车间调度问题[J]. 计算机应用, 2021, 41(8): 2249-2257.
[2]	张盟, 郭健全. 需求和回收不确定的闭环供应链渠道结构选择[J]. 计算机应用, 2021, 41(7): 2100-2107.
[3]	杨震, 马健霄, 王宝杰. 设置待行区条件下双环相位信号配时优化模型[J]. 计算机应用, 2021, 41(7): 2108-2112.
[4]	李进, 王凤, 杨沈宇. 换电模式下电动车货运路径优化模型与算法[J]. 计算机应用, 2021, 41(6): 1792-1798.
[5]	黄梨, 卢龙. 基于长距离依赖编码与深度残差U-Net的缺血性卒中病灶分割[J]. 计算机应用, 2021, 41(6): 1820-1827.
[6]	李舒仪, 韩晓龙. 海铁联运港口混合作业模式下轨道吊与集卡协同调度[J]. 计算机应用, 2021, 41(5): 1506-1513.
[7]	孙鹤立, 孙玉柱, 张晓云. 基于生成对抗网络的事件描述生成[J]. 计算机应用, 2021, 41(5): 1256-1261.
[8]	周美玲, 陈淮莉. 基于负荷平衡的电动汽车模糊多目标充电调度算法[J]. 计算机应用, 2021, 41(4): 1192-1198.
[9]	王彬溶, 谭代伦, 郑伯川. 基于旅行商问题转化和遗传算法求解汽配件喷涂顺序[J]. 计算机应用, 2021, 41(3): 881-886.
[10]	马晓梅, 何非. 基于改进遗传算法的标签印刷生产调度技术[J]. 计算机应用, 2021, 41(3): 860-866.
[11]	黄书召, 田军委, 乔路, 王沁, 苏宇. 基于改进遗传算法的无人机路径规划[J]. 计算机应用, 2021, 41(2): 390-397.
[12]	屠红艳, 张挺, 夏鹏飞, 杜奕. 基于改进型变分自编码器的不确定性空间信息重建方法[J]. 计算机应用, 2021, 41(10): 2959-2963.
[13]	杨华龙, 王美玉, 辛禹辰. 横向整合战略下异质车辆库存路径优化模型[J]. 计算机应用, 2021, 41(10): 3040-3048.
[14]	陈桢, 钟一文, 林娟. 求解0-1背包问题的混合贪婪遗传算法[J]. 计算机应用, 2021, 41(1): 87-94.
[15]	王永航, 张天宇, 郑红星. 考虑恶劣天气的班轮多阶段重调度方法[J]. 计算机应用, 2021, 41(1): 286-294.