基于相对极差的不确定脑网络特征提取与分类

doi:10.3778/j.issn.1002-8331.2101-0350

摘要/Abstract

摘要： 近年来，脑网络被广泛应用在脑部疾病的诊断和分类中。考虑到大脑的不确定性特征，先前的研究将不确定图应用在脑网络建模中。在不确定脑网络研究中，传统的特征提取方法多采用均值、方差、极差等进行子图特征提取，但是它们存在泛化性能差以及子图间差异无法直接衡量的问题，进而影响分类准确率。因此，相对极差被提出作为新的特征提取方法。这一方法的优点在于既考虑到子图模式间的最大差异，又考虑到子图模式间的组间差异，可以有效避免传统方法的弊端。结果表明，相对极差与其他特征提取方法相比，其分类性能显著高于传统方法。同时，在不同的特征选择方法下相对极差表现出较好的分类性能，具有很强的泛化性。该研究为不确定脑网络特征提取方法提供了重要的参考意义。

关键词: 不确定脑网络, 频繁子图, 特征选择, 机器学习, 分类

Abstract: In recent years, brain network has been widely used in the diagnosis and classification of brain diseases. Considering the uncertainty characteristics of brain, some researches applied uncertain graph to brain network modeling. In the research of uncertain brain networks, traditional feature extraction methods mostly used mean, variance, range and other methods to extract subgraph features, but they have the problems of poor generalization performance and the difference between subgraphs cannot be directly measured. Therefore, relative range is proposed as the new feature extraction method. The advantage of this method is that it not only takes into account the maximum differences between subgraph patterns, but also takes into account the differences between subgraph patterns, which can effectively avoid the disadvantages of traditional methods. The results show that compared with other feature extraction methods, the classification performance of the proposed method is significantly higher than that of the traditional method. Meanwhile, under different feature selection methods, the relative range shows good classification performance and strong generalization. This study provides an important reference for feature extraction of uncertain brain networks.

Key words: uncertain brain network, frequent subgraphs, feature selection, machine learning, classification

孙超, 闻敏, 李鹏祖, 李瑶, Ibegbu Nnamdi JULIAN, 郭浩. 基于相对极差的不确定脑网络特征提取与分类[J]. 计算机工程与应用, 2022, 58(14): 126-133.

SUN Chao, WEN Min, LI Pengzu, LI Yao, Ibegbu Nnamdi JULIAN, GUO Hao. Feature Extraction and Classification of Uncertain Brain Network Based on Relative Range[J]. Computer Engineering and Applications, 2022, 58(14): 126-133.

参考文献

[1] LIN H，CAI X，ZHANG D，et al.Functional connectivity markers of depression in advanced Parkinson’s disease[J].NeuroImage：Clinical，2019，25：1-7.
[2] CUI X H，XIANG J，GUO H，et al.Classification of Alzheimer’s disease，mild cognitive impairment，and normal controls with subnetwork selection and graph kernel principal component analysis based on minimum spanning tree brain functional network[J].Frontiers in Computational Neuroscience，2018，12（31）：1-12.
[3] GARRISON K A，SCHEINOST D，FINNE S，et al.The （in）stability of functional brain network measures across thresholds[J].Neuroimage，2015，118（15）：651-661.
[4] KONG X，YU P S，WANG X，et al.Discriminative feature selection for uncertain graph classification[J].Cognitive Computation，2013，7：346-358.
[5] MORRISS J，GELL M，REEKUM C M V.The uncertain brain：a co-ordinate based meta-analysis of the neural signatures supporting uncertainty during different contexts[J].Neuroence & Biobehavioral Reviews，2019，96：241-249.
[6] MICHAEL D R，KARSTEN K，JEAN Y，et al.An uncertainty visual analytics framework for fMRI functional connectivity[J].Neuroinformatics，2018，10：1-13.
[7] YE Y，GUOREN，WANG L，et al.Efficient pattern matching on big uncertain graphs[J].Information Sciences，2016，339：369-394.
[8] KHAN A，YE Y，CHEN L.On uncertain graphs[J].Synthesis Lectures on Data Management，2018，10（1）：1-94.
[9] DU L，LI C，CHEN H，et al.Probabilistic SimRank computation over uncertain graphs[J].Information Sciences，2015，295（6）：521-535.
[10] ZHU R，ZOU Z，LI J.Towards efficient top-k reliability search on uncertain graphs[J].Knowledge & Information Systems，2017，50（3）：1-28.
[11] 孙欢欢，郭浩，陈俊杰.面向脑网络的图挖掘及特征选择方法[J].计算机工程与设计，2018，39（9）：2912-2917.
SUN H H，GUO H，CHEN J J.Graph mining and feature selection for brain network[J].Computer Engineering and Design，2018，39（9）：2912-2917.
[12] 刘峰，GODFRED K M，李欣芸，等.不确定脑网络的异常拓扑分析及分类研究[J].计算机工程与应用，2020，56（2）：127-132.
LIU F，GODFRED K M，LI X Y，et al.Abnormal topological analysis and classification research of uncertain brain networks[J].Computer Engineering and Applications，2020，56（2）：127-132.
[13] CHEN E J.Range statistics and equivalence tests[J].Journal of Simulation，2014，8（2）：143-150.
[14] MATHEW T.The analysis of variance[J].Technometrics，2018，43（2）：233-234.
[15] VERGARA V M，MAYER A R，DAMARAJU E，et al.The effect of preprocessing in dynamic functional network connectivity used to classify mild traumatic brain injury[J].Brain & Behavior，2017，21：1-11.
[16] LIN C H，CHI C Y，CHEN L，et al.Detection of sources in non-negative blind source separation by minimum description length criterion[J].IEEE Transactions on Neural Networks & Learning Systems，2017，99：1-16.
[17] 茂旭，杨剑，杨阳.基于独立成分分析功能连接的抑郁症分类研究[J].计算机应用研究，2018，35（6）：1641-1644.
MAO X，YANG J，YANG Y.Independent component analysis based functional connectivity for classification of depression[J].Application Research of Computers，2018，35（6）：1641-1644.
[18] CAI X，XIE D，MADSEN K H，et al.Generalizability of machine learning for classification of schizophrenia based on resting-state functional MRI data[J].Human Brain Mapping，2020，41（6）：172-184.
[19] SALIMI-KHORSHIDI G，DOUAUD G，BECKMANN C F，et al.Automatic denoising of functional MRI data：combining independent component analysis and hierarchical fusion of classifiers[J].Neuroimage，2014，90（8）：449-468.
[20] GUO H，YAN P P，CHENG C，et al.fMRI classification method with multiple feature fusion based on minimum spanning tree analysis[J].Psychiatry Research Neuroimaging，2018，277：14-27.
[21] 邹兆年，李建中，高宏，等.从不确定图中挖掘频繁子图模式[J].软件学报，2009，20（11）：2965-2976.
ZOU Z N，LI J Z，GAO H，et al.Mining frequent subgraph patterns from uncertain graphs[J].Journal of Software，2009，20（11）：2965-2976.
[22] 韩蒙，张炜，李建中.RAKING：一种高效的不确定图K-极大频繁模式挖掘算法[J].计算机学报，2010，33（8）：1387-1395.
HAN M，ZHANG W，LI J Z.RAKING：an efficient K-maximal frequent pattern mining algorithm on uncertain graph database[J].Chinese Journal of Computers，2010，33（8）：1387-1395.
[23] CHEN Y，ZHAO X，LIN X，et al.Efficient mining of frequent patterns on uncertain graphs[J].IEEE Transactions on Knowledge & Data Engineering，2019，31：287-300.
[24] JIN N，WANG W.LTS：discriminative subgraph mining by learning from search history[C]//Proceedings of the 27th International Conference on Data Engineering，2011：207-218.
[25] CSABA H，KINGA-OIGA R，ROSIAN G，et al.Assessing landslide vulnerability using bivariate statistical analysis and the frequency ratio model.Case study：Transylvanian Plain （Romania）[J].Ztschrift Fur Geomorphologie，2016，60（4）：359-371.
[26] AHAD N A，ALIPIAH F M，AZHARI F.Applicability of G-test in analyzing categorical variables[C]//4th Innovation and Analytics Conference & Exhibition（IACE 2019），2019：1-6.
[27] TIAN X，HE J，SHI Y.Statistical dependence test with Hilbert-Schmidt independence criterion[J].Journal of Physics：Conference Series，2020，1601：1-8.
[28] 梁礼明，陈明理，刘博文，等.基于图论的支持向量机核函数选择[J].计算机工程与设计，2019，40（5）：1316-1321.
LIANG L M，CHEN M L，LIU B W，et al.Selection of SVM kernel function based on graph theory[J].Computer Engineering and Design，2019，40（5）：1316-1321.
[29] SHAIKH S，CHANGAN L，MALIK M R，et al.Software defect-prone classification using machine learning：a virtual classification study between LibSVM & LibLinear[C]//2019 13th International Conference on Mathematics，Actuarial Science，Computer Science and Statistics（MACS），2020：1-6.
[30] 张俐，王枞.基于最大相关最小冗余联合互信息的多标签特征选择算法[J].通信学报，2018，39（5）：111-122.
ZHANG L，WANG C.Multi-label feature selection algorithm based on joint mutual information of max-relevance and min-redundancy[J].Journal on Communications，2018，39（5）：111-122.
[31] 朱雪玲，袁福来.基于区域的抑郁症默认网络内部功能连接研究[J].中国临床心理学杂志，2016，24（2）：218-220.
ZHU X L，YUAN F L.Default mode network in depressive patients based on region-wise functional connecti-
vity[J].Chinese Journal of Clinical Psychology，2016，24（2）：218-220.
[32] DANIEL R，MATTEO M，PAOLA M，et al.Opposing changes in the functional architecture of large-scale networks in bipolar mania and depression[J].Schizophrenia Bulletin，2020，4：4-14.
[33] WANG H Y，MO C，FANG F.Dissociated deficits in attentional networks in social anxiety and depression[J].Science China（Life Sciences），2020，63（7）：131-138.