Classifying brain states and determining the discriminating activation patterns: Support Vector Machine on functional MRI data

Mourão-Miranda, Janaı́na; Bokde, Arun L.W.; Born, Christine; Hampel, Harald; Stetter, M.

doi:10.1016/j.neuroimage.2005.06.070

Cited by 641 publications

(551 citation statements)

References 27 publications

Supporting

Mentioning

544

Contrasting

Order By: Relevance

“…The function that predicts patient or control status from data is called a classifier. The parameters of the function can be learned from data by using a large number of well-known statistical machine learning methods, including (among others) support vector machines (Mourao-Miranda et al, 2005). The classification problem can be seen as a high-dimensional learning task, because there are many more features (dimensions) in the vector space than training samples.…”

Section: Classification Techniquementioning

confidence: 99%

“…Finally, we use a non-parametric permutation testing approach to decide which features should be considered as having a significant discriminative weight (Nichols and Holmes, 2002;Mourao-Miranda et al, 2005). To this end, we generate 20 random permutations of the class labels (control or patient), and run cross-validation experiments.…”

Section: Computation Of Discriminative Weightsmentioning

confidence: 99%

“…Recent work has highlighted the feasibility of using single structural scans for reliable MS diagnosis (Rovira et al, 2009), the ability of local multivariate predictive methods to discriminate between MS patients and controls with high accuracy, even when using NABT structural data (Weygandt et al, 2011), and the possibility of using global multivariate methods with structural data to distinguish various aspects of MS severity (Bendfeldt et al, 2012). Accordingly, given the increased use and development of predictive modelling techniques in fMRI research, originally derived from machine learning or pattern recognition (Kamitani and Tong, 2005;Mourao-Miranda et al, 2005;Ethofer et al, 2009;Weil and Rees, 2010;Shirer et al, 2011), it would appear highly suitable and advantageous to apply similar techniques to characterise high-dimensional fMRI data obtained during restingstate (Richiardi et al, 2010(Richiardi et al, , 2011. There has also been a slow concurrent increase in the use of multivariate predictive modelling techniques applied to functional connectivity data of pathological subjects.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Classifying minimally disabled multiple sclerosis patients from resting state functional connectivity

et al. 2012

View full text Add to dashboard Cite

Section: Classification Techniquementioning

confidence: 99%

Section: Computation Of Discriminative Weightsmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Classifying minimally disabled multiple sclerosis patients from resting state functional connectivity

et al. 2012

View full text Add to dashboard Cite

“…Two classification algorithms, classification trees and Fisher linear discriminative analysis (FLDA), were used in order to validate the classification accuracy. FLDA has been widely used in fMRI-based pattern recognition (e.g., Carlson et al, 2003;LaConte et al, 2003;Mourao-Miranda et al, 2005). FLDA works most efficiently if the smallest group has significantly more cases than the number of variables and when groups are approximately of equal size.…”

Section: Classification Analysismentioning

confidence: 99%

The neural substrates for atypical planning and execution of word production in stuttering

Chen

Ning

et al. 2010

Experimental Neurology

View full text Add to dashboard Cite

a b s t r a c t a r t i c l e i n f oUsing an fMRI-based classification approach and the structural equation modeling (SEM) method, this study examined the neural bases of atypical planning and execution processes involved in stuttering. Twelve stuttering speakers and 12 controls were asked to name pictures under different conditions (single-syllable, multi-syllable, or repeated-syllable) in the scanner. The contrasts between conditions provided information about planning and execution processes. The classification analysis showed that, as compared to nonstuttering controls, stuttering speakers' atypical planning of speech was evident in their neural activities in the bilateral inferior frontal gyrus (IFG) and right putamen and their atypical execution of speech was evident in their activations in the right cerebellum and insula, left premotor area (PMA), and angular gyrus (AG). SEM results further revealed two parallel neural circuits-the basal ganglia-IFG/PMA circuit and the cerebellum-PMA circuit-that were involved in atypical planning and execution processes of stuttering, respectively. The AG appeared to be involved in the interface of atypical planning and execution in stuttering. These results are discussed in terms of their implications to the theories about stuttering and to clinical applications.

show abstract

“…It has been noted [Cox and Savoy, 2003] that this correspondence makes fMRI well suited for ''brain reading'' experiments, based on the modality's noninvasiveness and spatiotemporal qualities. While many fundamental methodological studies of fMRI classification exist Kustra and Strother, 2001;LaConte et al, 2003LaConte et al, , 2005bMartinez-Ramon et al, 2006;Mitchell et al, 2004;Mourao-Miranda et al, 2005;Shaw et al, 2003;Strother et al, 2002Strother et al, , 2004, much interest in predicting brain states and studying mental representations was catalyzed by the desire to evaluate the evidence for a localized versus distributed coding scheme for the (high-order) extrastriate visual cortex [Cox and Savoy, 2003;Downing et al, 2001;Hanson et al, 2004;Haxby et al, 2001;Ishai et al, 1999;Kanwisher et al, 1997;. Consequently, there has been a remarkable surge in cognitive neuroscientific interest and inventive experimental designs focused on classification of brain states from fMRI data.…”

Section: Introductionmentioning

confidence: 99%

Real‐time fMRI using brain‐state classification

LaConte¹,

Peltier²,

Hu³

2006

Human Brain Mapping

224

187

View full text Add to dashboard Cite

Abstract:We have implemented a real-time functional magnetic resonance imaging system based on multivariate classification. This approach is distinctly different from spatially localized real-time implementations, since it does not require prior assumptions about functional localization and individual performance strategies, and has the ability to provide feedback based on intuitive translations of brain state rather than localized fluctuations. Thus this approach provides the capability for a new class of experimental designs in which real-time feedback control of the stimulus is possible-rather than using a fixed paradigm, experiments can adaptively evolve as subjects receive brain-state feedback. In this report, we describe our implementation and characterize its performance capabilities. We observed $80% classification accuracy using whole brain, block-design, motor data. Within both left and right motor task conditions, important differences exist between the initial transient period produced by task switching (changing between rapid left or right index finger button presses) and the subsequent stable period during sustained activity. Further analysis revealed that very high accuracy is achievable during stable task periods, and that the responsiveness of the classifier to changes in task condition can be much faster than signal time-to-peak rates. Finally, we demonstrate the versatility of this implementation with respect to behavioral task, suggesting that our results are applicable across a spectrum of cognitive domains. Beyond basic research, this technology can complement electroencephalography-based brain computer interface research, and has potential applications in the areas of biofeedback rehabilitation, lie detection, learning studies, virtual reality-based training, and enhanced conscious awareness.

show abstract

Classifying brain states and determining the discriminating activation patterns: Support Vector Machine on functional MRI data

Cited by 641 publications

References 27 publications

Classifying minimally disabled multiple sclerosis patients from resting state functional connectivity

Classifying minimally disabled multiple sclerosis patients from resting state functional connectivity

The neural substrates for atypical planning and execution of word production in stuttering

Real‐time fMRI using brain‐state classification

Contact Info

Product

Resources

About