Elena S. Ackley

Enhancement of Temporal Resolution and BOLD Sensitivity in Real-Time fMRI using Multi-Slab Echo-Volumar Imaging

In this study, a new approach to high-speed fMRI using multi-slab echo-volumar imaging (EVI) is developed that minimizes geometrical image distortion and spatial blurring, and enables nonaliased sampling of physiological signal fluctuation to increase BOLD sensitivity compared to conventional echo-planar imaging (EPI). Real-time fMRI using whole brain 4-slab EVI with 286 ms temporal resolution (4mm isotropic voxel size) and partial brain 2-slab EVI with 136 ms temporal resolution (4×4×6 mm(3) voxel size) was performed on a clinical 3 Tesla MRI scanner equipped with 12-channel head coil. Four-slab EVI of visual and motor tasks significantly increased mean (visual: 96%, motor: 66%) and maximum t-score (visual: 263%, motor: 124%) and mean (visual: 59%, motor: 131%) and maximum (visual: 29%, motor: 67%) BOLD signal amplitude compared with EPI. Time domain moving average filtering (2s width) to suppress physiological noise from cardiac and respiratory fluctuations further improved mean (visual: 196%, motor: 140%) and maximum (visual: 384%, motor: 200%) t-scores and increased extents of activation (visual: 73%, motor: 70%) compared to EPI. Similar sensitivity enhancement, which is attributed to high sampling rate at only moderately reduced temporal signal-to-noise ratio (mean: -52%) and longer sampling of the BOLD effect in the echo-time domain compared to EPI, was measured in auditory cortex. Two-slab EVI further improved temporal resolution for measuring task-related activation and enabled mapping of five major resting state networks (RSNs) in individual subjects in 5 min scans. The bilateral sensorimotor, the default mode and the occipital RSNs were detectable in time frames as short as 75 s. In conclusion, the high sampling rate of real-time multi-slab EVI significantly improves sensitivity for studying the temporal dynamics of hemodynamic responses and for characterizing functional networks at high field strength in short measurement times.

Online Negative Databases

The benefits of negative detection for obscuring information are explored in the context of Artificial Immune Systems (AIS). AIS based on string matching have the potential for an extra security feature in which the “normal” profile of a system is hidden from its possible hijackers. Even if the model of normal behavior falls into the wrong hands, reconstructing the set of valid or “normal” strings is an \(\mathcal{NP}\)-hard problem. The data-hiding aspects of negative detection are explored in the context of an application to negative databases. Previous work is reviewed describing possible representations and reversibility properties for privacy-enhancing negative databases. New algorithms are described, which allow on-line creation and updates of negative databases, and future challenges are discussed.

High-speed real-time resting-state FMRI using multi-slab echo-volumar imaging.

We recently demonstrated that ultra-high-speed real-time fMRI using multi-slab echo-volumar imaging (MEVI) significantly increases sensitivity for mapping task-related activation and resting-state networks (RSNs) compared to echo-planar imaging (Posse et al., 2012). In the present study we characterize the sensitivity of MEVI for mapping RSN connectivity dynamics, comparing independent component analysis (ICA) and a novel seed-based connectivity analysis (SBCA) that combines sliding-window correlation analysis with meta-statistics. This SBCA approach is shown to minimize the effects of confounds, such as movement, and CSF and white matter signal changes, and enables real-time monitoring of RSN dynamics at time scales of tens of seconds. We demonstrate highly sensitive mapping of eloquent cortex in the vicinity of brain tumors and arterio-venous malformations, and detection of abnormal resting-state connectivity in epilepsy. In patients with motor impairment, resting-state fMRI provided focal localization of sensorimotor cortex compared with more diffuse activation in task-based fMRI. The fast acquisition speed of MEVI enabled segregation of cardiac-related signal pulsation using ICA, which revealed distinct regional differences in pulsation amplitude and waveform, elevated signal pulsation in patients with arterio-venous malformations and a trend toward reduced pulsatility in gray matter of patients compared with healthy controls. Mapping cardiac pulsation in cortical gray matter may carry important functional information that distinguishes healthy from diseased tissue vasculature. This novel fMRI methodology is particularly promising for mapping eloquent cortex in patients with neurological disease, having variable degree of cooperation in task-based fMRI. In conclusion, ultra-high-real-time speed fMRI enhances the sensitivity of mapping the dynamics of resting-state connectivity and cerebro-vascular pulsatility for clinical and neuroscience research applications.

Protecting data privacy through hard-to-reverse negative databases

The paper extends the idea of negative representations of information for enhancing privacy. Simply put, a set DB of data elements can be represented in terms of its complement set. That is, all the elements not in DB are depicted and DB itself is not explicitly stored. review the negative database (NDB) representation scheme for storing a negative image compactly and propose a design for depicting a multiple record DB using a collection of NDBs—in contrast to the single NDB approach of previous work. Finally, we present a method for creating negative databases that are hard to reverse in practice, i.e., from which it is hard to obtain DB, by adapting a technique for generating 3-SAT formulas.

Amygdala Regulation Following fMRI-Neurofeedback without Instructed Strategies

Within the field of functional magnetic resonance imaging (fMRI) neurofeedback, most studies provide subjects with instructions or suggest strategies to regulate a particular brain area, while other neuro-/biofeedback approaches often do not. This study is the first to investigate the hypothesis that subjects are able to utilize fMRI neurofeedback to learn to differentially modulate the fMRI signal from the bilateral amygdala congruent with the prescribed regulation direction without an instructed or suggested strategy and apply what they learned even when feedback is no longer available. Thirty-two subjects were included in the analysis. Data were collected at 3 Tesla using blood oxygenation level dependent (BOLD)-sensitivity optimized multi-echo EPI. Based on the mean contrast between up- and down-regulation in the amygdala in a post-training scan without feedback following three neurofeedback sessions, subjects were able to regulate their amygdala congruent with the prescribed directions with a moderate effect size of Cohen’s d = 0.43 (95% conf. int. 0.23–0.64). This effect size would be reduced, however, through stricter exclusion criteria for subjects that show alterations in respiration. Regulation capacity was positively correlated with subjective arousal ratings and negatively correlated with agreeableness and susceptibility to anger. A learning effect over the training sessions was only observed with end-of-block feedback (EoBF) but not with continuous feedback (trend). The results confirm the above hypothesis. Further studies are needed to compare effect sizes of regulation capacity for approaches with and without instructed strategies.

Spatially aggregated multiclass pattern classification in functional MRI using optimally selected functional brain areas.

In previous works, boosting aggregation of classifier outputs from discrete brain areas has been demonstrated to reduce dimensionality and improve the robustness and accuracy of functional magnetic resonance imaging (fMRI) classification. However, dimensionality reduction and classification of mixed activation patterns of multiple classes remain challenging. In the present study, the goals were (a) to reduce dimensionality by combining feature reduction at the voxel level and backward elimination of optimally aggregated classifiers at the region level, (b) to compare region selection for spatially aggregated classification using boosting and partial least squares regression methods and (c) to resolve mixed activation patterns using probabilistic prediction of individual tasks. Brain activation maps from interleaved visual, motor, auditory and cognitive tasks were segmented into 144 functional regions. Feature selection reduced the number of feature voxels by more than 50%, leaving 95 regions. The two aggregation approaches further reduced the number of regions to 30, resulting in more than 75% reduction of classification time and misclassification rates of less than 3%. Boosting and partial least squares (PLS) were compared to select the most discriminative and the most task correlated regions, respectively. Successful task prediction in mixed activation patterns was feasible within the first block of task activation in real-time fMRI experiments. This methodology is suitable for sparsifying activation patterns in real-time fMRI and for neurofeedback from distributed networks of brain activation.

Negative Ternary Set-Sharing

The Set-Sharing domain has been widely used to infer at compile-time interesting properties of logic programs such as occurs-check reduction, automatic parallelization, and finite-tree analysis. However, performing abstract unification in this domain requires a closure operation that increases the number of sharing groups exponentially. Much attention has been given to mitigating this key inefficiency in this otherwise very useful domain. In this paper we present a novel approach to Set-Sharing: we define a new representation that leverages the complement (or negative) sharing relationships of the original sharing set, without loss of accuracy. Intuitively, given an abstract state

Diffusion tensor spectroscopic imaging of the human brain in children and adults

sh_{\mathcal{V}}

On-line Negative Databases.

over the finite set of variables of interest

Real-time Functional MRI with Multi-Echo EPI and Spatially Aggregated Pattern Classification

\mathcal{V}