EVALUATING STATIONARITY VIA CHANGE-POINT ALTERNATIVES WITH APPLICATIONS TO FMRI DATA

被引:68
作者
Aston, John A. D. [1 ]
Kirch, Claudia [2 ]
机构
[1] Univ Warwick, Dept Stat, CRiSM, Coventry CV4 7AL, W Midlands, England
[2] KIT, Inst Stochast, D-76133 Karlsruhe, Germany
基金
英国工程与自然科学研究理事会;
关键词
Epidemic change; functional time series; high-dimensional data; resting state fMRI; separable covariance structure; stationarity; PRINCIPAL-COMPONENT ANALYSIS; STATISTICAL-ANALYSIS; BRAIN; SPACE; CONNECTIVITY; SEPARABILITY;
D O I
10.1214/12-AOAS565
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Functionalmagnetic resonance imaging (fMRI) is now a well-established technique for studying the brain. However, in many situations, such as when data are acquired in a resting state, it is difficult to know whether the data are truly stationary or if level shifts have occurred. To this end, change-point detection in sequences of functional data is examined where the functional observations are dependent and where the distributions of change-points from multiple subjects are required. Of particular interest is the case where the change-point is an epidemic change-a change occurs and then the observations return to baseline at a later time. The case where the covariance can be decomposed as a tensor product is considered with particular attention to the power analysis for detection. This is of interest in the application to fMRI, where the estimation of a full covariance structure for the three-dimensional image is not computationally feasible. Using the developed methods, a large study of resting state fMRI data is conducted to determine whether the subjects undertaking the resting scan have nonstationarities present in their time courses. It is found that a sizeable proportion of the subjects studied are not stationary. The change-point distribution for those subjects is empirically determined, as well as its theoretical properties examined.
引用
收藏
页码:1906 / 1948
页数:43
相关论文
共 51 条
[31]   Improved optimization for the robust and accurate linear registration and motion correction of brain images [J].
Jenkinson, M ;
Bannister, P ;
Brady, M ;
Smith, S .
NEUROIMAGE, 2002, 17 (02) :825-841
[32]   Block permutation principles for the change analysis of dependent data [J].
Kirch, Claudia .
JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2007, 137 (07) :2453-2474
[33]   TFT-BOOTSTRAP: RESAMPLING TIME SERIES IN THE FREQUENCY DOMAIN TO OBTAIN REPLICATES IN THE TIME DOMAIN [J].
Kirch, Claudia ;
Politis, Dimitris N. .
ANNALS OF STATISTICS, 2011, 39 (03) :1427-1470
[34]   Change-point in the mean of dependent observations [J].
Kokoszka, P ;
Leipus, R .
STATISTICS & PROBABILITY LETTERS, 1998, 40 (04) :385-393
[35]  
Lahiri SN., 2003, Resampling Methods for Dependent Data
[36]   Modeling state-related fMRI activity using change-point theory [J].
Lindquist, Martin A. ;
Waugh, Christian ;
Wager, Tor D. .
NEUROIMAGE, 2007, 35 (03) :1125-1141
[37]   STATE-SPACE SOLUTIONS TO THE DYNAMIC MAGNETOENCEPHALOGRAPHY INVERSE PROBLEM USING HIGH PERFORMANCE COMPUTING [J].
Long, Christopher J. ;
Purdon, Patrick L. ;
Temereanca, Simona ;
Desai, Neil U. ;
Haemaelaeinen, Matti S. ;
Brown, Emery N. .
ANNALS OF APPLIED STATISTICS, 2011, 5 (2B) :1207-1228
[38]   Testing for separability of space-time covariances [J].
Mitchell, MW ;
Genton, MG ;
Gumpertz, ML .
ENVIRONMETRICS, 2005, 16 (08) :819-831
[39]   AUTOMATED ANALYSIS OF QUANTITATIVE IMAGE DATA USING ISOMORPHIC FUNCTIONAL MIXED MODELS, WITH APPLICATION TO PROTEOMICS DATA [J].
Morris, Jeffrey S. ;
Baladandayuthapani, Veerabhadran ;
Herrick, Richard C. ;
Sanna, Pietro ;
Gutstein, Howard .
ANNALS OF APPLIED STATISTICS, 2011, 5 (2A) :894-923
[40]   Quantifying the uncertainty in change points [J].
Nam, Christopher F. H. ;
Aston, John A. D. ;
Johansen, Adam M. .
JOURNAL OF TIME SERIES ANALYSIS, 2012, 33 (05) :807-823