Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (8): 2288-2293.DOI: 10.11772/j.issn.1001-9081.2020101553

Special Issue: 多媒体计算与计算机仿真

• Multimedia computing and computer simulation • Previous Articles     Next Articles

Classification of functional magnetic resonance imaging data based on semi-supervised feature selection by spectral clustering

ZHU Cheng, ZHAO Xiaoqi, ZHAO Liping, JIAO Yuhong, ZHU Yafei, CHENG Jianying, ZHOU Wei, TAN Ying   

  1. The Key Laboratory for Computer Systems of State Ethnic Affairs Commission(Southwest Minzu University), Chengdu Sichuan 610041, China
  • Received:2020-10-10 Revised:2021-01-12 Online:2021-08-10 Published:2021-01-27
  • Supported by:
    This work is partially supported by the Sichuan Science and Technology Project (2019YFF0207, 2019YFH0055), the 2018-2020 Sichuan Higher Education Talent Training Quality and Teaching Reform Project (JG2018-312), the Fundamental Research Funds for the Central Universities (2016NGJPY06).

基于谱聚类半监督特征选择的功能磁共振成像数据分类

祝承, 赵晓琦, 赵丽萍, 焦玉宏, 朱亚飞, 陈建英, 周伟, 谭颖   

  1. 计算机系统国家民委重点实验室(西南民族大学), 成都 610041
  • 通讯作者: 谭颖
  • 作者简介:祝承(1997-),男,浙江杭州人,硕士研究生,主要研究方向:智能信息处理;赵晓琦(1997-),女,辽宁铁岭人,硕士研究生,主要研究方向:智能信息处理;赵丽萍(1997-),女,四川南充人,硕士研究生,主要研究方向:智能信息处理;焦玉宏(1996-),女,甘肃白银人,硕士研究生,主要研究方向:神经网络分析及信息处理;朱亚飞(2000-),男,河南濮阳人,主要研究方向:深度学习;陈建英(1970-),女,四川冕宁人,教授,博士,主要研究方向:数据仓库、数据挖掘;周伟(1977-),男,浙江温州人,副教授,博士,主要研究方向:神经网络、机器学习;谭颖(1974-),男,四川成都人,教授,博士,主要研究方向:人工智能、医学图像分析。
  • 基金资助:
    四川省科技项目(2019YFG0207,2019YFH0055);四川省2018—2020年高等教育人才培养质量和教学改革项目(JG2018-312);中央高校基本科研业务费专项(2016NGJPY06)。

Abstract: Aiming at the high-dimensional and small sample problems of functional Magnetic Resonance Imaging (fMRI) data, a Semi-Supervised Feature Selection by Spectral Clustering (SS-FSSC) model was proposed. Firstly, the prior brain region template was used to extract the time series signal. Then, the Pearson correlation coefficient and the Order Statistics Correlation Coefficient (OSCC) were selected to describe the functional connection features between the brain regions, and spectral clustering was performed to the features. Finally, the feature importance criterion based on Constraint score was adopted to select feature subsets, and the subsets were input into the Support Vector Machine (SVM) classifier for classification. By 100 times of five-fold cross-validation on the COBRE (Center for Biomedical Research Excellence) schizophrenia public dataset in the experiments, it is found that when the number of retained features is 152, the highest average accuracy of the proposed model to schizophrenia is about 77%, and the highest accuracy of the proposed model to schizophrenia is 95.83%. Experimental result analysis shows that by only retaining 16 functional connection features for classifier training, the model can stably achieve an average accuracy of more than 70%. In addition, in the results obtained by the proposed model, Intracalcarine Cortex has the highest occurrence frequency among the 10 brain regions corresponding to the functional connections, which is consistent to the existing research state about schizophrenia.

Key words: functional Magnetic Resonance Imaging (fMRI), feature selection by spectral clustering, Pearson correlation coefficient, Order Statistics Correlation Coefficient (OSCC), schizophrenia, Intracalcarine Cortex

摘要: 针对功能磁共振成像(fMRI)数据的高维度小样本问题,提出谱聚类半监督特征选择(SS-FSSC)模型。首先利用先验脑区模板提取时间序列信号;然后选取皮尔逊相关系数与序统计量相关系数(OSCC)描述脑区间的功能连接特征,并对特征进行谱聚类;最后利用基于Constraint得分的特征重要性准则挑选出特征子集,并把这些子集输入支持向量机(SVM)分类器进行分类。实验通过在COBRE精神分裂症公开数据集上重复进行100次五折交叉验证,发现当保留特征数为152时,所提模型对精神分裂症得到最高平均准确率约为77%,最高准确率为95.83%。实验结果分析表明,模型仅保留16个功能连接特征进行分类器训练就能稳定达到70%以上的平均准确率,且所提模型得到的结果中功能连接对应的10个脑区中距状裂皮质(Intracalcarine Cortex)出现频次最高,符合现有对精神分裂症的研究状况。

关键词: 功能磁共振成像, 谱聚类特征选择, 皮尔逊相关系数, 序统计量相关系数, 精神分裂症, 距状裂皮质

CLC Number: