Stephen M. LaConte

Support vector machines for temporal classification of block design fMRI data

This paper treats support vector machine (SVM) classification applied to block design fMRI, extending our previous work with linear discriminant analysis [LaConte, S., Anderson, J., Muley, S., Ashe, J., Frutiger, S., Rehm, K., Hansen, L.K., Yacoub, E., Hu, X., Rottenberg, D., Strother S., 2003a. The evaluation of preprocessing choices in single-subject BOLD fMRI using NPAIRS performance metrics. NeuroImage 18, 10-27; Strother, S.C., Anderson, J., Hansen, L.K., Kjems, U., Kustra, R., Siditis, J., Frutiger, S., Muley, S., LaConte, S., Rottenberg, D. 2002. The quantitative evaluation of functional neuroimaging experiments: the NPAIRS data analysis framework. NeuroImage 15, 747-771]. We compare SVM to canonical variates analysis (CVA) by examining the relative sensitivity of each method to ten combinations of preprocessing choices consisting of spatial smoothing, temporal detrending, and motion correction. Important to the discussion are the issues of classification performance, model interpretation, and validation in the context of fMRI. As the SVM has many unique properties, we examine the interpretation of support vector models with respect to neuroimaging data. We propose four methods for extracting activation maps from SVM models, and we examine one of these in detail. For both CVA and SVM, we have classified individual time samples of whole brain data, with TRs of roughly 4 s, thirty slices, and nearly 30,000 brain voxels, with no averaging of scans or prior feature selection.

The Quantitative Evaluation of Functional Neuroimaging Experiments: The NPAIRS Data Analysis Framework

We introduce a data-analysis framework and performance metrics for evaluating and optimizing the interaction between activation tasks, experimental designs, and the methodological choices and tools for data acquisition, preprocessing, data analysis, and extraction of statistical parametric maps (SPMs). Our NPAIRS (nonparametric prediction, activation, influence, and reproducibility resampling) framework provides an alternative to simulations and ROC curves by using real PET and fMRI data sets to examine the relationship between prediction accuracy and the signal-to-noise ratios (SNRs) associated with reproducible SPMs. Using cross-validation resampling we plot training-test set predictions of the experimental design variables (e.g., brain-state labels) versus reproducibility SNR metrics for the associated SPMs. We demonstrate the utility of this framework across the wide range of performance metrics obtained from [(15)O]water PET studies of 12 age- and sex-matched data sets performing different motor tasks (8 subjects/set). For the 12 data sets we apply NPAIRS with both univariate and multivariate data-analysis approaches to: (1) demonstrate that this framework may be used to obtain reproducible SPMs from any data-analysis approach on a common Z-score scale (rSPM[Z]); (2) demonstrate that the histogram of a rSPM[Z] image may be modeled as the sum of a data-analysis-dependent noise distribution and a task-dependent, Gaussian signal distribution that scales monotonically with our reproducibility performance metric; (3) explore the relation between prediction and reproducibility performance metrics with an emphasis on bias-variance tradeoffs for flexible, multivariate models; and (4) measure the broad range of reproducibility SNRs and the significant influence of individual subjects. A companion paper describes learning curves for four of these 12 data sets, which describe an alternative mutual-information prediction metric and NPAIRS reproducibility as a function of training-set sizes from 2 to 18 subjects. We propose the NPAIRS framework as a validation tool for testing and optimizing methodological choices and tools in functional neuroimaging.

Real-time fMRI neurofeedback: progress and challenges.

In February of 2012, the first international conference on real time functional magnetic resonance imaging (rtfMRI) neurofeedback was held at the Swiss Federal Institute of Technology Zurich (ETHZ), Switzerland. This review summarizes progress in the field, introduces current debates, elucidates open questions, and offers viewpoints derived from the conference. The review offers perspectives on study design, scientific and clinical applications, rtfMRI learning mechanisms and future outlook.

Real-Time fMRI Using Brain-State Classification

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. Hum Brain Mapp 2006.

Medical Image Processing on the GPU - Past, Present and Future

Graphics processing units (GPUs) are used today in a wide range of applications, mainly because they can dramatically accelerate parallel computing, are affordable and energy efficient. In the field of medical imaging, GPUs are in some cases crucial for enabling practical use of computationally demanding algorithms. This review presents the past and present work on GPU accelerated medical image processing, and is meant to serve as an overview and introduction to existing GPU implementations. The review covers GPU acceleration of basic image processing operations (filtering, interpolation, histogram estimation and distance transforms), the most commonly used algorithms in medical imaging (image registration, image segmentation and image denoising) and algorithms that are specific to individual modalities (CT, PET, SPECT, MRI, fMRI, DTI, ultrasound, optical imaging and microscopy). The review ends by highlighting some future possibilities and challenges.

Neuroimaging after mild traumatic brain injury: review and meta-analysis.

This paper broadly reviews the study of mild traumatic brain injury (mTBI), across the spectrum of neuroimaging modalities. Among the range of imaging methods, however, magnetic resonance imaging (MRI) is unique in its applicability to studying both structure and function. Thus we additionally performed meta-analyses of MRI results to examine 1) the issue of anatomical variability and consistency for functional MRI (fMRI) findings, 2) the analogous issue of anatomical consistency for white-matter findings, and 3) the importance of accounting for the time post injury in diffusion weighted imaging reports. As we discuss, the human neuroimaging literature consists of both small and large studies spanning acute to chronic time points that have examined both structural and functional changes with mTBI, using virtually every available medical imaging modality. Two key commonalities have been used across the majority of imaging studies. The first is the comparison between mTBI and control populations. The second is the attempt to link imaging results with neuropsychological assessments. Our fMRI meta-analysis demonstrates a frontal vulnerability to mTBI, demonstrated by decreased signal in prefrontal cortex compared to controls. This vulnerability is further highlighted by examining the frequency of reported mTBI white matter anisotropy, in which we show a strong anterior-to-posterior gradient (with anterior regions being more frequently reported in mTBI). Our final DTI meta-analysis examines a debated topic arising from inconsistent anisotropy findings across studies. Our results support the hypothesis that acute mTBI is associated with elevated anisotropy values and chronic mTBI complaints are correlated with depressed anisotropy. Thus, this review and set of meta-analyses demonstrate several important points about the ongoing use of neuroimaging to understand the functional and structural changes that occur throughout the time course of mTBI recovery. Based on the complexity of mTBI, however, much more work in this area is required to characterize injury mechanisms and recovery factors and to achieve clinically-relevant capabilities for diagnosis.

An open science resource for establishing reliability and reproducibility in functional connectomics.

Efforts to identify meaningful functional imaging-based biomarkers are limited by the ability to reliably characterize inter-individual differences in human brain function. Although a growing number of connectomics-based measures are reported to have moderate to high test-retest reliability, the variability in data acquisition, experimental designs, and analytic methods precludes the ability to generalize results. The Consortium for Reliability and Reproducibility (CoRR) is working to address this challenge and establish test-retest reliability as a minimum standard for methods development in functional connectomics. Specifically, CoRR has aggregated 1,629 typical individuals’ resting state fMRI (rfMRI) data (5,093 rfMRI scans) from 18 international sites, and is openly sharing them via the International Data-sharing Neuroimaging Initiative (INDI). To allow researchers to generate various estimates of reliability and reproducibility, a variety of data acquisition procedures and experimental designs are included. Similarly, to enable users to assess the impact of commonly encountered artifacts (for example, motion) on characterizations of inter-individual variation, datasets of varying quality are included.

Activity and effective connectivity of parietal and occipital cortical regions during haptic shape perception

It is now widely accepted that visual cortical areas are active during normal tactile perception, but the underlying mechanisms are still not clear. The goal of the present study was to use functional magnetic resonance imaging (fMRI) to investigate the activity and effective connectivity of parietal and occipital cortical areas during haptic shape perception, with a view to potentially clarifying the role of top-down and bottom-up inputs into visual areas. Subjects underwent fMRI scanning while engaging in discrimination of haptic shape or texture, and in separate runs, visual shape or texture. Accuracy did not differ significantly between tasks. Haptic shape-selective regions, identified on a contrast between the haptic shape and texture conditions in individual subjects, were found bilaterally in the postcentral sulcus (PCS), multiple parts of the intraparietal sulcus (IPS) and the lateral occipital complex (LOC). The IPS and LOC foci tended to be shape-selective in the visual modality as well. Structural equation modelling was used to study the effective connectivity among the haptic shape-selective regions in the left hemisphere, contralateral to the stimulated hand. All possible models were tested for their fit to the correlations among the observed time-courses of activity. Two equivalent models emerged as the winners. These models, which were quite similar, were characterized by both bottom-up paths from the PCS to parts of the IPS, and top-down paths from the LOC and parts of the IPS to the PCS. We conclude that interactions between unisensory and multisensory cortical areas involve bidirectional information flow.

Decoding fMRI brain states in real-time.

This article reviews a technological advance that originates from two areas of ongoing neuroimaging innovation-(1) the use of multivariate supervised learning to decode brain states and (2) real-time functional magnetic resonance imaging (rtfMRI). The approach uses multivariate methods to train a model capable of decoding a subjects brain state from fMRI images. The decoded brain states can be used as a control signal for a brain computer interface (BCI) or to provide neurofeedback to the subject. The ability to adapt the stimulus during the fMRI experiment adds a new level of flexibility for task paradigms and has potential applications in a number of areas, including performance enhancement, rehabilitation, and therapy. Multivariate approaches to real-time fMRI are complementary to region-of-interest (ROI)-based methods and provide a principled method for dealing with distributed patterns of brain responses. Specifically, a multivariate approach is advantageous when network activity is expected, when mental strategies could vary from individual to individual, or when one or a few ROIs are not unequivocally the most appropriate for the investigation. Beyond highlighting important developments in rtfMRI and supervised learning, the article discusses important practical issues, including implementation considerations, existing resources, and future challenges and opportunities. Some possible future directions are described, calling for advances arising from increased experimental flexibility, improvements in predictive modeling, better comparisons across rtfMRI and other BCI implementations, and further investigation of the types of feedback and degree to which interface modulation is obtainable for various tasks.

Comparison of α-chloralose, medetomidine and isoflurane anesthesia for functional connectivity mapping in the rat

Functional connectivity measures based upon low-frequency blood-oxygenation-level-dependent functional magnetic resonance imaging (BOLD fMRI) signal fluctuations have become a widely used tool for investigating spontaneous brain activity in humans. Still unknown, however, is the precise relationship between neural activity, the hemodynamic response and fluctuations in the MRI signal. Recent work from several groups had shown that correlated low-frequency fluctuations in the BOLD signal can be detected in the anesthetized rat - a first step toward elucidating this relationship. Building on this preliminary work, through this study, we demonstrate that functional connectivity observed in the rat depends strongly on the type of anesthesia used. Power spectra of spontaneous fluctuations and the cross-correlation-based connectivity maps from rats anesthetized with alpha-chloralose, medetomidine or isoflurane are presented using a high-temporal-resolution imaging sequence that ensures minimal contamination from physiological noise. The results show less localized correlation in rats anesthetized with isoflurane as compared with rats anesthetized with alpha-chloralose or medetomidine. These experiments highlight the utility of using different types of anesthesia to explore the fundamental physiological relationships of the BOLD signal and suggest that the mechanisms contributing to functional connectivity involve a complicated relationship between changes in neural activity, neurovascular coupling and vascular reactivity.