Decoding brain states from fMRI connectivity graphs
Research highlights
► Whole-brain functional connectivity can be used for inter-subject decoding. ► Embedding connectivity graphs in vector space is effective for graph matching. ► Ensembling techniques help discrimination, both within and between frequency subbands.
Introduction
Traditional fMRI analysis consists of univariate statistical hypothesis testing to assess changes in the activity of each brain voxel induced by the stimulation paradigm (Frackowiak et al., 1997). More recently, approaches derived from supervised machine learning – commonly termed “brain decoding” in the field of neuroimaging – have shown that it is possible to exploit more subtle relationships in voxels' intensity patterns (Haxby et al., 2001, Haynes & Rees, 2006, Norman et al., 2006). These methods rely on a classifier to predict the subject's brain state from the BOLD responses in a set of selected voxels, such as visual (Cox & Savoy, 2003, Haynes & Rees, 2005, Kamitani & Tong, 2005, Thirion et al., 2006, Kay et al., 2008, Miyawaki et al., 2008) or auditory cortices (Ethofer et al., 2009). Results from brain decoding are often remarkable since they clearly reach beyond the possibilities of univariate techniques, but also because they are able to uncover information from fine-grained cortical activity despite the relatively low spatial resolution of fMRI. Furthermore, instead of a fixed spatial window, one can also apply classification to a so-called “searchlight” that slides over the whole-brain data, such that the classification success for each position of the spotlight can then be mapped to show brain regions that carry discriminative information between different conditions (Kriegeskorte et al., 2006). Another interesting approach is to use spatiotemporal observations as an input to the classifier (Mitchell et al., 2004, Mourao-Miranda et al., 2007).
The study of functional connectivity is concerned with the temporal coherence between neurophysiological events observed in spatially remote brain regions. In early work, correlation with a seed voxel was investigated and revealed bilateral coactivation between sensory cortices (Biswal et al., 1995, Lowe et al., 1998). Further advances have been driven by unsupervised methods such as source separation – mainly principal components (Friston et al., 1993) and independent components analysis (McKeown et al., 1998, Calhoun et al., 2002, Beckmann & Smith, 2004) which allow to identify large-scale cortical networks – and by other methods such as dynamic causal modelling, which tries to establish effective connectivity and requires prior information about the neurological network to investigate Friston et al. 2003. A recent method related to our work proposed to use resting-state correlations between regions of interest as features for an SVM classifier (Craddock et al., 2009). Another attractive methodology to investigate functional networks is to rely on mathematical graph theory; i.e., constructing the (undirected) graph from temporal correlation matrices and computing related measures; e.g., node degree, hubbiness, and so on (Sporns et al., 2000, Salvador et al., 2005). This methodology has brought new insights in functional connectivity at resting state, such as the small-world organisation of cortical networks at low temporal frequencies (Achard et al., 2006).
Here, we bring together brain decoding and graph representations based on functional connectivity measures. These measures, such as temporal correlation, are performed over a given period of time and reflect the timecourse resemblances between different regions. Moreover, we estimate connectivity at different temporal scales using the wavelet transform as a preprocessing step. Then, we build a classifier trained on functional connectivity graphs of a group of subjects to distinguish between different brain states of an unseen subject. The aim of our approach is to identify the connections that are most discriminative between brain states, and to obtain relevant visual representation of the data for neuroscience studies.
Previous brain decoding techniques primarily rely on linear support vector machines (SVMs), which use a soft-margin hyperplane to separate classes. In the present work, we propose polythetic decision trees that fit a hyperplane using the most discriminative features (i.e., connections) at each level, such that potentially complex and non-linear class boundaries can be obtained by multilevel trees. Efficient and effective learning of this type of decision trees relies on recent advances in pattern recognition (Friedman et al., 2000, Gama, 2004, Landwehr et al., 2005) and provides embedded feature selection which yields a compact discriminant function whose parameters are amenable to interpretation. Their variance properties make them good candidates for ensembling, steering the classification strategy towards slightly weaker but simpler (lower capacity) classifiers, which is a desirable behaviour in high-dimensional learning, a classical situation in fMRI where the number of dimensions is much higher than the number of training examples.
Our paper is organised as follows. In the Methods section, we describe our data processing pipeline together with the details of the proposed methodology. Next, we illustrate the feasibility of our approach by a proof-of-concept; i.e., an fMRI experiment with block-based stimulation paradigm (watching short movies) with long resting periods. We extract the set of connections that is the most discriminative between rest and stimulation at different temporal scales. The method is able to correctly classify the conditions in a leave-one-subject-out cross-validation setting. Interestingly, we find that the low-frequency correlations of the BOLD signal (below 0.11 Hz) are the most informative. The discriminative network also confirms a differential modulation of sensory areas, in particular within the visual system, and midline brain areas during movie and rest conditions.
Section snippets
Preprocessing and data representation
The preprocessing steps are illustrated schematically in Fig. 1, and explained in detail below.
After realignment of the functional volumes using SPM5,1 we use the IBASPM toolbox (Tzourio-Mazoyer et al., 2002, Alemán-Gómez et al., 2006) to build an individual brain atlas based on the structural MRI, containing M = 90 anatomical regions. While this is a relatively coarse atlas, it is an essential step to allow for inter-subject variability and enable
Subjects and data acquisition
The N = 15 subjects (4 males, 11 females) were aged between 18 and 36 years old, without history of neurological disorders. They had given written informed consent to participate in the study, which was performed in accordance with the local Ethics Committee of the University of Geneva. Scanning was performed on a Siemens 3 T Tim Trio. Functional imaging data were acquired in two sessions using gradient-echo echo-planar imaging (TR/TE/FA = 1.1 s/27 ms/90°, matrix = 64 × 64, voxel size = 3.75 × 3.75 × 4.2 mm3, 21
Classification
Classifier training and testing are performed using the leave-one-subject-out procedure outlined in Classification section of Methods. Stratified classification results are given in Table 1 (column L = 1, α = 100%): in each of the 15 cross-validation folds, 2 tests are performed, one for the resting and one for the movies condition of a single subject. Thus, the granularity of results is about 3%. One striking result is that lower-frequency subbands (3 and 4) have much more discriminative power
Classification
The relatively high number of connections that are retained by statistical feature selection (Graph embedding and feature selection) in the low-frequency subbands, with respect to the number retained in higher-frequency subbands, hints at the presence of resting-state networks that are consistent across subjects (Damoiseaux et al., 2006, Mantini et al., 2007), which in turn yields relatively low inter-subject standard deviation on the weights of graph edges that these networks comprise. Not
Conclusion
In summary, we have proposed a classification approach to infer brain states from functional connectivity graphs, instead of the commonly used brain voxel activation values. We have shown that the approach is applicable to inter-subject brain decoding with good results, and that interpretable output can be generated. We have demonstrated the feasibility using a cognitive task and compared the discriminative connectivity graph with SPM-style activation patterns. The potential of the proposed
Acknowledgments
The authors wish to thank the anonymous reviewers for their valuable input. This work was supported in part by the Swiss National Science Foundation (grant PP00P2-123438), in part by the Société Académique de Genève and the FOREMANE foundation, and in part by the Center for Biomedical Imaging (CIBM) of the Geneva and Lausanne Universities, EPFL, and the Leenaards and Louis-Jeantet foundations.
References (73)
- et al.
Classification of spatially unaligned fMRI scans
Neuroimage
(2010) - et al.
Functional magnetic resonance imaging (fMRI) “brain reading”: detecting and classifying distributed patterns of fMRI activity in human visual cortex
Neuroimage
(2003) - et al.
Combining multivariate voxel selection and support vector machines for mapping and classification of fMRI spatial patterns
Neuroimage
(2008) - et al.
Decoding of emotional information in voice-sensitive cortices
Current Biology
(2009) - et al.
Dynamic causal modeling
Neuroimage
(2003) - et al.
Visual image reconstruction from human brain activity using a combination of multiscale local image decoders
Neuron
(2008) - et al.
Beyond mind-reading: multivoxel pattern analysis of fMRI data
Trends in Cognitive Sciences
(2006) - et al.
Connectivity and complexity: the relationship between neuroanatomy and brain dynamics
Neural Networks
(2000) - et al.
Inverse retinotopy: inferring the visual content of images from brain activation patterns
Neuroimage
(2006) - et al.
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
Neuroimage
(2002)
Whole-brain anatomical networks: does the choice of nodes matter?
Neuroimage
A resilient, low-frequency, small-world human brain functional network with highly connected association cortical hubs
The Journal of Neuroscience
IBASPM: toolbox for automatic parcellation of brain structures
A block spin construction of ondelettes. Part I: Lemarié functions
Communications in Mathematical Physics
Probabilistic independent component analysis for functional magnetic resonance imaging
IEEE Transactions on Medical Imaging
Functional connectivity in the motor cortex of resting human brain using echo-planar MRI
Magnetic Resonance in Medicine
Majority system and the Condorcet jury theorem
Statistician
Classification and Regression Trees
Independent component analysis of fMRI data in the complex domain
Magnetic Resonance in Medicine
Thirty years of graph matching in pattern recognition
International Journal of Pattern Recognition and Artificial Intelligence
Consistent resting-state networks across healthy subjects
Proceedings of the National Academy of Science
Pattern Classification
Human Brain Function
Cited by (251)
Deep learning models of cognitive processes constrained by human brain connectomes
2022, Medical Image AnalysisAdvancing brain network models to reconcile functional neuroimaging and clinical research
2022, NeuroImage: Clinical