计算机科学 ›› 2022, Vol. 49 ›› Issue (8): 257-266.doi: 10.11896/jsjkx.210600094

• 人工智能 • 上一篇    下一篇

基于多尺度的稀疏脑功能超网络构建及多特征融合分类研究

李瑶1, 李涛1, 李埼钒1, 梁家瑞2, Ibegbu Nnamdi JULIAN1, 陈俊杰1, 郭浩1   

  1. 1 太原理工大学信息与计算机学院 山西 晋中 030600
    2 太原理工大学软件学院 山西 晋中 030600
  • 收稿日期:2021-06-10 修回日期:2021-10-13 发布日期:2022-08-02
  • 通讯作者: 郭浩(feiyu_guo@sina.com)
  • 作者简介:(1813672649@qq.com)
  • 基金资助:
    国家自然科学基金(61472270,61672374,61741212,61876124,61976150,61873178);山西省研究生教育创新项目(2020BY131);山西省重点研发计划(201803D31043);山西省科技厅应用基础研究项目青年面上项目(201801D121135,201803D31043)

Construction and Multi-feature Fusion Classification Research Based on Multi-scale Sparse Brain Functional Hyper-network

LI Yao1, LI Tao1, LI Qi-fan1, LIANG Jia-rui2, Ibegbu Nnamdi JULIAN1, CHEN Jun-jie1, GUO Hao1   

  1. 1 College of Information and Computer,Taiyuan University of Technology,Jinzhong,Shanxi 030600,China
    2 College of Software,Taiyuan University of Technology,Jinzhong,Shanxi 030600,China
  • Received:2021-06-10 Revised:2021-10-13 Published:2022-08-02
  • About author:LI Yao,born in 1996,postgraduate.Her main research interests include artificial intelligence,intelligent information processing and brain imaging.
    GUO Hao,Ph.D,professor,is a senior member of China Computer Federation.His main research interests include artificial intelligence,brain imaging and medical big data research.
  • Supported by:
    National Natural Science Foundation of China(61472270,61672374,61741212,61876124,61976150,61873178),Shanxi Province Graduate Education Innovation Project(2020BY131),Key Research and Development Project of Shanxi Province(201803D31043) and Youth General Project of Shanxi Provincial Department of Science and Technology(201801D121135,201803D31043).

摘要: 脑功能超网络已成功应用于脑疾病的诊断。在之前的研究中,集中通过改变超边的方法来改善超网络的构建,忽略了不同尺度的节点定义对脑功能超网络拓扑的影响。考虑到该问题,提出了基于不同尺度的脑区划分来进行脑功能超网络的创建,从而分析其对脑功能超网络拓扑和分类性能的影响。具体来说,首先,基于自动解剖标记模板,利用聚类算法和随机动态种子点的方法对大脑进行细分割;其次,基于每种节点规模下所得的平均时间序列,利用LASSO方法分别进行脑功能超网络的构建;接着分别提取功能超网络的多组局部特征(节点度、最短路径长度、聚类系数),并利用非参数检验和基于相关的方法选取每种节点规模下的差异特征;最后,分别利用支持向量机构建分类模型。分类结果显示,随着节点规模的增大,所构建的脑功能超网络分类准确率增高,在节点尺度1501下,准确率高达95.45%。同时,多尺度融合的分类准确率优于任一尺度下的分类准确率,这表明不同尺度的节点定义会影响脑功能超网络的拓扑,在未来的脑功能超网络研究中,除了关注超边的构建方法外,应更加关注大脑划分方案的选择,而且多种基于大脑划分的尺度融合特征可以补充更多的分类信息,提高抑郁症与正常人的分类性能。此外,无论是在哪种节点规模下,多组局部属性特征的分类性能均优于单一属性的分类性能。该结果表明,多组局部属性特征可以弥补单一特征的缺失信息,从而发现更多的脑疾病生物学标志物,在有效表征脑功能超网络模型的同时,还需要多角度地量化脑功能超网络拓扑信息,这样才可增强组间差异表征能力,提高脑疾病诊断的预测能力。

关键词: 多尺度, 机器学习, 静息态功能磁共振成像, 局部属性特征, 脑功能超网络, 抑郁症

Abstract: Brain functional hypernetworks have been successfully utilized in the diagnosis of brain diseases.In the previous study,the different hyper-edge generation method was mainly used to improve the construction of the hyper-network,which ignored the influence of different nodes definitions on the brain functional hyper-network topology.Therefore,in light of this problem,it is proposed to construct a brain functional hyper-network based on parcellation of different scales,so as to analyze its impact on brain functional hyper-network topology and classification performance.Specifically,firstly,based on the anatomical automatic labeling atlas,the brain was segmented by the method of clustering algorithm and the random dynamic seed point;secondly,based on the average time series obtained under each node scale,the brain functional hyper-network was constructed by the LASSO method respectively;then multiple sets of local features (node degree,shortest path,clustering coefficient) were extracted,and non-parametric tests and correlation-based methods were used to select features with difference;finally,support vector machine was adopted to build classification model.The classification results show that as the size of nodes increases,the classification accuracy of the constructed brain functional hyper-network is higher.When the node scale is 1501,the classification accuracy can reach 95.45%.Meanwhile,the classification accuracy of multi-scale fusion is better than that of any scale,which indicate different node definitions will affect the topology of the brain functional hyper-network.In future research,besides focusing to the construction method of the hyper-edge,the choice of brain parcellation scheme needs more attention in hyper-network.Moreover,combining multi-scale features can supplement more classification information to enhance the classification performance of depression and normal control.In addition,regardless of the size of the node,the classification performance of multiple sets of local properties is better than that of a single type of properties,which illustrates multiple sets of local property can make up for the missing information of a type of single feature,thereby discovering more brain disease biological markers.While effectively representing the brain functional hyper-network,it is also necessary to quantify the brain functional hyper-network topology information from multiple angles,so that the ability to characterize differences between groups can be enhanced,and the ability to diagnose and predict brain diseases can be improved.

Key words: Brain functional hyper-network, Depression, Local property feature, Machine learning, Multi-scale, Resting state functional magnetic resonance imaging

中图分类号: 

  • TP181
[1]MORGAN S E,ACHARD S,TERMENON M,et al.Low-di-mensional morpho-space of topological motifs in human fMRI brain networks[J].Network Neuroscience,2018,2(2):285-302.
[2]NI H,LI W,CHEN X.Brain network evolution modeling based on Alzheimer’s disease[J].Chinese Journal of Intelligent Science and Technology,2019,1(4):369-378.
[3]ZHANG Q,WANG B.Progress in the study of functional magnetic resonance imaging (fMRI) brain networks in the depression[J].Chinese Journal of Magnetic Resonance Imaging,2018,9(4):289-293.
[4]WU X J.Research on the Neural Mechanism of Face Recognition Based on the Analysis of the Functional Connectivity of the Resting Brain[D].Beijing:Beijing Jiaotong University,2018.
[5]ZHANG D Q,ZHU Q,HAO X K,et al.Intelligent analysis of brain images[J].Scientia Sinica(Informationis),2018,48(5):109-122.
[6]LI Y,LIU J Y,GAO X Q,et al.Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification[J].Medical Image Analysis,2018,52:80-96.
[7]BARAVALLE R,MONTANI F.Higher-Order CumulantsDrive Neuronal Activity Patterns,Inducing UP-DOWN States in Neural Populations[J].Entropy,2020,22(4):477.
[8]GLICKFELD L L,OLSEN S R.Higher-Order Areas of theMouse Visual Cortex[J].Annual Review of Vision Science,2017,3(1):251-273.
[9]MONTANGIE L,MONTANI F.Higher-order correlations incommon input shapes the output spiking activity of a neural population[J].Physica A:Statistical Mechanics and its Applications,2017,471:845-861.
[10]JIE B,WEE C Y,SHEN D,et al.Hyper-connectivity of functional networks for brain disease diagnosis[J].Medical Image Analysis,2016,32:84.
[11]BATTISTON F,CENCETTI G,C I I,et al.Networks beyond pairwise interactions:Structure and dynamics[J].Physics Reports,2020,874:1-92.
[12]LI Y,GAO X Q,JIE B,et al.Multimodal Hyper-connectivityNetworks for MCI Classification[C]//Medical Image Computing and Computer Assisted Intervention(MICCAI).2017:433-441.
[13]LI Y,SUN C,LI P Z,et al.Hypernetwork Construction andFeature Fusion Analysis Based on Sparse Group Lasso Method on fMRI Dataset[J/OL].Frontiers in Neuroscience,2020,14(60).https://doi.org/10.3389/fnins.2020.00060.
[14]GUO H,LI Y,XU Y,et al.Resting-State Brain Functional Hyper-Network Construction Based on Elastic Net and Group Lasso Methods[J].Frontiers in Neuroinformatics,2018,12:25.
[15]YOUSRA A,BASIT R,KAMRAN M A,et al.A multi-modal,multi-atlas-based approach for Alzheimer detection via machine learning[J].International Journal of Imaging Systems and Technology,2018,28(2):113-123.
[16]JING B,LONG Z Q,LIU H,et al.Identifying current and remitted major depressive disorder with the Hurst exponent:a comparative study on two automated anatomical labeling atlases[J].Oncotarget,2017,8(52):90452-90464.
[17]LONG Z,HUANG J,LI B,et al.A Comparative Atlas-Based Recognition of Mild Cognitive Impairment with Voxel-Based Morphometry[J/OL].Frontiers in Neuroscience,2018,12(916).https://doi.org/10.3389/fnins.2018.00916.
[18]HU Y,WANG L J,NIE S D.Review on brain functional parcellation based on resting-state functional magnetic resonance imaging data[J].Journal of Image and Graphics,2017,22(10):1325-1334.
[19]XIAO Q.A Method for Measuring Node Importance in Hypernetwork Model[J].Research Journal of Applied Sciences,Engineering and Technology,2013,5(2):568-573.
[20]MA T,SUO Q.Review of hypernetwork based on Hypergraph[J].Operations Research and Management Science,2021,30(2):232-239.
[21]HEINTZ B,HONG R,SINGH S,et al.MESH:A Flexible Distributed Hypergraph Processing System[C]//IEEE Internatio-nal Conference on Cloud Engineering (IC2E).Prague:IEEE Press,2019:12-22.
[22]DUAN T,LI X B,ZHOU X,et al.Review on the research progress of super-network:theory,application and tools[C]//The Conference of China System Simulation Technology and Application (CCSSTA).Xinjiang,2019.
[23]KAUFMANN M,KREVELD M,SPECKMANN B.Subdivision Drawings of Hypergraphs[C]//International Symposium on Graph Drawing.2016:396-407.
[24]MAZOYER N T,LANDEAU B,PAPATHANASSIOU D,et al.Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain[J].Neuroimage,2002,15(1):273-289.
[25]PARK H S,JUN C H.A simple and fast algorithm for K-medoids clustering[J].Expert Systems with Applications,2009,36(2):3336-3341.
[26]GUO H.Machine learning classifier using abnormal resting state functional brain netwrok topological metrics in major depressive disorder[D].Taiyuan:TaiYuan University of Technology,2013.
[27]LI Y,GAO X,JIE B,et al.Multimodal Hyper-connectivity Networks for MCI Classification[C]//Medical Image Computing and Computer Assisted Intervention(MICCAI 2017).2017:433-441.
[28]ZU C,GAO Y,MUNSELL B,et al.Identifying disease-relatedsubnetwork connectome biomarkers by sparse hypergraph lear-ning[J].Brain Imaging and Behavior,2018,13:879-892.
[29]PENG Y,ZU C,ZHANG D Q.Hypergraph Based Multi-Modal Feature Selection and Its Application[J].Journal of Frontiers of Computer Science and Technology,2018,12(1):112-119.
[30]LI C X,HAO X K,ZHANG D Q.Hyper-Network Guided Correlation Analysis on Imaging Genetics[J].Pattern Recognition and Artificial Intelligence,2017,30(9):841-849.
[31]LI Y,LIU J Y,GAO X Q,et al.Multimodal hyper-connectivity of functional networks using functionally-weighted LASSO for MCI classification[J].Medical Image Analysis,2019,52:80-96.
[32]ZHANG Z Z,LIN H J,GAO Y.Dynamic Hypergraph Structure Learning[C]//Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence.2018:3162-3169.
[33]LIU Q S,SUN Y B,WANG C T,et al.Elastic Net Hypergraph Learning for Image Clustering and Semi-supervised Classification[J].IEEE Transactions on Image Processing,2016,26(1):452-463.
[34]GUO H,ZHANG F,CHEN J J,et al.Machine Learning Classification Combining Multiple Features of A Hyper-Network of fMRI Data in Alzheimer’s Disease[J/OL].Frontiers in neuroscience,2017,11(615).https://doi.org/10.3389/fnins.2017.00615.
[35]WANG M L,HAO X K,HUANG J S,et al.Discovering network phenotype between genetic risk factors and disease status via diagnosis-aligned multi-modality regression method in Alzhei-mer’s disease[J].Bioinformatics,2018,35(11):1948-1957.
[36]LIU J,JI S W,YE J P.SLEP:Sparse Learning with Efficient Projections[J/OL].Arizona State University,2013.http://www.public.asu.edu/~jye02/Software/SLEP.
[37]BRIER M R,THOMAS J B,FAGAN A M,et al.Functionalconnectivity and graph theory in preclinical Alzheimer’s disease[J].Neurobiology of Aging,2014,35(4):757-768.
[38]XIAO Q,YANG F L,LUO S,et al.A Node Importance Measu-ring Method based on Hypernetwork[C]//International Confe-rence on Future Generation Communication and Networking.2016:187-196.
[39]ZHANG Z K,LIU C.A hypergraph model of social tagging networks[J/OL].Journal of Statistical Mechanics:Theory and Experiment,2010,10:10005.https://iopscience.iop.org/article/10.1088/1742-5468/2010/10/P10005/meta.
[40]GALLAGHER S R,GOLDBERG D S.Clustering Coefficients in Protein Interaction Hypernetworks[C]//Proceedings of the International Conference on Bioinformatics,Computational Biology and Biomedical Informatics.2013:552-560.
[41]FASANO G,FRANCESCHINI A.A multidimensional version of the Kolmogorov-Smirnov test[J].Monthly Notices of the Royal Astronomical Society,1987,50(1):9-20.
[42]BENJAMINI Y,HOCHBERG Y.Controlling The False Disco-very Rate-A Practical And Powerful Approach To Multiple Testing[J].Journal of the Royal Statistical Society,1995,57(57):289-300.
[43]YANG J,YANG J Y,ZHANG D,et al.Feature fusion:parallel strategy vs.serial strategy[J].Pattern Recognition,2003,36(6):1369-1381.
[44]YU W,LIU T,VALDEZ R,et al.Application of support vector machine modeling for prediction of common diseases:the case of diabetes and pre-diabetes[J/OL].BMC Medical Informatics and Decision Making,2010,10(1):16.https://doi.org/10.1186/1472-6947-10-16.
[45]DASGUPTA S,GOLDBERG Y,KOSOROK M.Feature elimination in kernel machines in moderately high dimensions[J].Annals of Statistics,2019,47:497-526.
[46]ZHAI J Q,YANG X X,CHENG Y Q,et al.Overview of Application of Fault Detection and Diagnosis Based on Machine Learning[J].Computer Measurement and Control,2021,29(3):1-9.
[47]WANG X,DONG Y Q,YU Q,et al.Review of Structural Support Vector Machines[J].Computer Engineering and Applications,2020,56(17):24-32.
[48]GUO H,LI Y,MENSAH G K,et al.Resting-State Functional Network Scale Effects and Statistical Significance-Based Feature Selection in Machine Learning Classification[J/OL].Computational and Mathematical Methods in Medicine,2019.https://doi.org/10.1155/2019/9108108.
[49]LV J L,JIANG X,LI X,et al.Sparse representation of whole-brain fMRI signals for identification of functional networks[J].Medical Image Analysis,2015,20(1):112-134.
[50]QIAO L,ZHANG H,KIM M,et al.Estimating functional brain networks by incorporating a modularity prior[J].NeuroImage,2016,141:399-407.
[51]LI X,WANG H.Identification of functional networks in restingstate fMRI data using adaptive sparse representation and affinity propagation clustering[J/OL].Frontiers in Neuroscience,2015,9(383).https://doi.org/10.3389/fnins.2015.00383.
[52]HSU C W,CHANG C C,LIN C J.A practical guide to support vector classification[OL].http://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf.Google Scholar.
[1] 冷典典, 杜鹏, 陈建廷, 向阳.
面向自动化集装箱码头的AGV行驶时间估计
Automated Container Terminal Oriented Travel Time Estimation of AGV
计算机科学, 2022, 49(9): 208-214. https://doi.org/10.11896/jsjkx.210700028
[2] 宁晗阳, 马苗, 杨波, 刘士昌.
密码学智能化研究进展与分析
Research Progress and Analysis on Intelligent Cryptology
计算机科学, 2022, 49(9): 288-296. https://doi.org/10.11896/jsjkx.220300053
[3] 何强, 尹震宇, 黄敏, 王兴伟, 王源田, 崔硕, 赵勇.
基于大数据的进化网络影响力分析研究综述
Survey of Influence Analysis of Evolutionary Network Based on Big Data
计算机科学, 2022, 49(8): 1-11. https://doi.org/10.11896/jsjkx.210700240
[4] 魏恺轩, 付莹.
基于重参数化多尺度融合网络的高效极暗光原始图像降噪
Re-parameterized Multi-scale Fusion Network for Efficient Extreme Low-light Raw Denoising
计算机科学, 2022, 49(8): 120-126. https://doi.org/10.11896/jsjkx.220200179
[5] 王馨彤, 王璇, 孙知信.
基于多尺度记忆残差网络的网络流量异常检测模型
Network Traffic Anomaly Detection Method Based on Multi-scale Memory Residual Network
计算机科学, 2022, 49(8): 314-322. https://doi.org/10.11896/jsjkx.220200011
[6] 张光华, 高天娇, 陈振国, 于乃文.
基于N-Gram静态分析技术的恶意软件分类研究
Study on Malware Classification Based on N-Gram Static Analysis Technology
计算机科学, 2022, 49(8): 336-343. https://doi.org/10.11896/jsjkx.210900203
[7] 杨炳新, 郭艳蓉, 郝世杰, 洪日昌.
基于数据增广和模型集成策略的图神经网络在抑郁症识别上的应用
Application of Graph Neural Network Based on Data Augmentation and Model Ensemble in Depression Recognition
计算机科学, 2022, 49(7): 57-63. https://doi.org/10.11896/jsjkx.210800070
[8] 陈明鑫, 张钧波, 李天瑞.
联邦学习攻防研究综述
Survey on Attacks and Defenses in Federated Learning
计算机科学, 2022, 49(7): 310-323. https://doi.org/10.11896/jsjkx.211000079
[9] 肖治鸿, 韩晔彤, 邹永攀.
基于多源数据和逻辑推理的行为识别技术研究
Study on Activity Recognition Based on Multi-source Data and Logical Reasoning
计算机科学, 2022, 49(6A): 397-406. https://doi.org/10.11896/jsjkx.210300270
[10] 姚烨, 朱怡安, 钱亮, 贾耀, 张黎翔, 刘瑞亮.
一种基于异质模型融合的 Android 终端恶意软件检测方法
Android Malware Detection Method Based on Heterogeneous Model Fusion
计算机科学, 2022, 49(6A): 508-515. https://doi.org/10.11896/jsjkx.210700103
[11] 孙福权, 崔志清, 邹彭, 张琨.
基于多尺度特征的脑肿瘤分割算法
Brain Tumor Segmentation Algorithm Based on Multi-scale Features
计算机科学, 2022, 49(6A): 12-16. https://doi.org/10.11896/jsjkx.210700217
[12] 李亚茹, 张宇来, 王佳晨.
面向超参数估计的贝叶斯优化方法综述
Survey on Bayesian Optimization Methods for Hyper-parameter Tuning
计算机科学, 2022, 49(6A): 86-92. https://doi.org/10.11896/jsjkx.210300208
[13] 赵璐, 袁立明, 郝琨.
多示例学习算法综述
Review of Multi-instance Learning Algorithms
计算机科学, 2022, 49(6A): 93-99. https://doi.org/10.11896/jsjkx.210500047
[14] 王飞, 黄涛, 杨晔.
基于Stacking多模型融合的IGBT器件寿命的机器学习预测算法研究
Study on Machine Learning Algorithms for Life Prediction of IGBT Devices Based on Stacking Multi-model Fusion
计算机科学, 2022, 49(6A): 784-789. https://doi.org/10.11896/jsjkx.210400030
[15] 方连花, 林玉梅, 吴伟志.
随机多尺度序决策系统的最优尺度选择
Optimal Scale Selection in Random Multi-scale Ordered Decision Systems
计算机科学, 2022, 49(6): 172-179. https://doi.org/10.11896/jsjkx.220200067
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!