João V. Duarte

Abnormal Brain Activation in Neurofibromatosis Type 1: A Link between Visual Processing and the Default Mode Network

Neurofibromatosis type 1 (NF1) is one of the most common single gene disorders affecting the human nervous system with a high incidence of cognitive deficits, particularly visuospatial. Nevertheless, neurophysiological alterations in low-level visual processing that could be relevant to explain the cognitive phenotype are poorly understood. Here we used functional magnetic resonance imaging (fMRI) to study early cortical visual pathways in children and adults with NF1. We employed two distinct stimulus types differing in contrast and spatial and temporal frequencies to evoke relatively different activation of the magnocellular (M) and parvocellular (P) pathways. Hemodynamic responses were investigated in retinotopically-defined regions V1, V2 and V3 and then over the acquired cortical volume. Relative to matched control subjects, patients with NF1 showed deficient activation of the low-level visual cortex to both stimulus types. Importantly, this finding was observed for children and adults with NF1, indicating that low-level visual processing deficits do not ameliorate with age. Moreover, only during M-biased stimulation patients with NF1 failed to deactivate or even activated anterior and posterior midline regions of the default mode network. The observation that the magnocellular visual pathway is impaired in NF1 in early visual processing and is specifically associated with a deficient deactivation of the default mode network may provide a neural explanation for high-order cognitive deficits present in NF1, particularly visuospatial and attentional. A link between magnocellular and default mode network processing may generalize to neuropsychiatric disorders where such deficits have been separately identified.

Multivariate pattern analysis reveals subtle brain anomalies relevant to the cognitive phenotype in neurofibromatosis type 1.

Neurofibromatosis Type 1 (NF1) is a common genetic condition associated with cognitive dysfunction. However, the pathophysiology of the NF1 cognitive deficits is not well understood. Abnormal brain structure, including increased total brain volume, white matter (WM) and grey matter (GM) abnormalities have been reported in the NF1 brain. These previous studies employed univariate model‐driven methods preventing detection of subtle and spatially distributed differences in brain anatomy. Multivariate pattern analysis allows the combination of information from multiple spatial locations yielding a discriminative power beyond that of single voxels. Here we investigated for the first time subtle anomalies in the NF1 brain, using a multivariate data‐driven classification approach. We used support vector machines (SVM) to classify whole‐brain GM and WM segments of structural T1‐weighted MRI scans from 39 participants with NF1 and 60 non‐affected individuals, divided in children/adolescents and adults groups. We also employed voxel‐based morphometry (VBM) as a univariate gold standard to study brain structural differences. SVM classifiers correctly classified 94% of cases (sensitivity 92%; specificity 96%) revealing the existence of brain structural anomalies that discriminate NF1 individuals from controls. Accordingly, VBM analysis revealed structural differences in agreement with the SVM weight maps representing the most relevant brain regions for group discrimination. These included the hippocampus, basal ganglia, thalamus, and visual cortex. This multivariate data‐driven analysis thus identified subtle anomalies in brain structure in the absence of visible pathology. Our results provide further insight into the neuroanatomical correlates of known features of the cognitive phenotype of NF1. Hum Brain Mapp 35:89–106, 2014.

Early disrupted neurovascular coupling and changed event level hemodynamic response function in type 2 diabetes: an fMRI study

Type 2 diabetes (T2DM) patients develop vascular complications and have increased risk for neurophysiological impairment. Vascular pathophysiology may alter the blood flow regulation in cerebral microvasculature, affecting neurovascular coupling. Reduced fMRI signal can result from decreased neuronal activation or disrupted neurovascular coupling. The uncertainty about pathophysiological mechanisms (neurodegenerative, vascular, or both) underlying brain function impairments remains. In this cross-sectional study, we investigated if the hemodynamic response function (HRF) in lesion-free brains of patients is altered by measuring BOLD (Blood Oxygenation Level-Dependent) response to visual motion stimuli. We used a standard block design to examine the BOLD response and an event-related deconvolution approach. Importantly, the latter allowed for the first time to directly extract the true shape of HRF without any assumption and probe neurovascular coupling, using performance-matched stimuli. We discovered a change in HRF in early stages of diabetes. T2DM patients show significantly different fMRI response profiles. Our visual paradigm therefore demonstrated impaired neurovascular coupling in intact brain tissue. This implies that functional studies in T2DM require the definition of HRF, only achievable with deconvolution in event-related experiments. Further investigation of the mechanisms underlying impaired neurovascular coupling is needed to understand and potentially prevent the progression of brain function decrements in diabetes.

Towards Personalized Neural Networks for Epileptic Seizure Prediction

Seizure prediction for untreatable epileptic patients, one of the major challenges of present neuroinformatics researchers, will allow a substantial improvement in their safety and quality of life. Neural networks, because of their plasticity and degrees of freedom, seem to be a good approach to consider the enormous variability of physiological systems. Several architectures and training algorithms are comparatively proposed in this work showing that it is possible to find an adequate network for one patient, but care must be taken to generalize to other patients. It is claimed that each patient will have his (her) own seizure prediction algorithms.

Parametric fMRI of paced motor responses uncovers novel whole-brain imaging biomarkers in spinocerebellar ataxia type 3.

Machado‐Joseph Disease, inherited type 3 spinocerebellar ataxia (SCA3), is the most common form worldwide. Neuroimaging and neuropathology have consistently demonstrated cerebellar alterations. Here we aimed to discover whole‐brain functional biomarkers, based on parametric performance‐level‐dependent signals. We assessed 13 patients with early SCA3 and 14 healthy participants. We used a combined parametric behavioral/functional neuroimaging design to investigate disease fingerprints, as a function of performance levels, coupled with structural MRI and voxel‐based morphometry. Functional magnetic resonance imaging (fMRI) was designed to parametrically analyze behavior and neural responses to audio‐paced bilateral thumb movements at temporal frequencies of 1, 3, and 5 Hz. Our performance‐level‐based design probing neuronal correlates of motor coordination enabled the discovery that neural activation and behavior show critical loss of parametric modulation specifically in SCA3, associated with frequency‐dependent cortico/subcortical activation/deactivation patterns. Cerebellar/cortical rate‐dependent dissociation patterns could clearly differentiate between groups irrespective of grey matter loss. Our findings suggest functional reorganization of the motor network and indicate a possible role of fMRI as a tool to monitor disease progression in SCA3. Accordingly, fMRI patterns proved to be potential biomarkers in early SCA3, as tested by receiver operating characteristic analysis of both behavior and neural activation at different frequencies. Discrimination analysis based on BOLD signal in response to the applied parametric finger‐tapping task significantly often reached >80% sensitivity and specificity in single regions‐of‐interest.Functional fingerprints based on cerebellar and cortical BOLD performance dependent signal modulation can thus be combined as diagnostic and/or therapeutic targets in hereditary ataxia. Hum Brain Mapp 37:3656–3668, 2016.

Pivotal role of hMT+ in long-range disambiguation of interhemispheric bistable surface motion: Long-Range Disambiguation of Ambiguous Surface Motion

It remains an open question whether long‐range disambiguation of ambiguous surface motion can be achieved in early visual cortex or instead in higher level regions, which concerns object/surface segmentation/integration mechanisms. We used a bistable moving stimulus that can be perceived as a pattern comprehending both visual hemi‐fields moving coherently downward or as two widely segregated nonoverlapping component objects (in each visual hemi‐field) moving separately inward. This paradigm requires long‐range integration across the vertical meridian leading to interhemispheric binding. Our fMRI study (n = 30) revealed a close relation between activity in hMT+ and perceptual switches involving interhemispheric segregation/integration of motion signals, crucially under nonlocal conditions where components do not overlap and belong to distinct hemispheres. Higher signal changes were found in hMT+ in response to spatially segregated component (incoherent) percepts than to pattern (coherent) percepts. This did not occur in early visual cortex, unlike apparent motion, which does not entail surface segmentation. We also identified a role for top–down mechanisms in state transitions. Deconvolution analysis of switch‐related changes revealed prefrontal, insula, and cingulate areas, with the right superior parietal lobule (SPL) being particularly involved. We observed that directed influences could emerge either from left or right hMT+ during bistable motion integration/segregation. SPL also exhibited significant directed functional connectivity with hMT+, during perceptual state maintenance (Granger causality analysis). Our results suggest that long‐range interhemispheric binding of ambiguous motion representations mainly reflect bottom–up processes from hMT+ during perceptual state maintenance. In contrast, state transitions maybe influenced by high‐level regions such as the SPL. Hum Brain Mapp 38:4882–4897, 2017.

Working memory load influences perceptual ambiguity by competing for fronto-parietal attentional resources

A visual stimulus is defined as ambiguous when observers perceive it as having at least two distinct and spontaneously alternating interpretations. Neuroimaging studies suggest an involvement of a right fronto-parietal network regulating the balance between stable percepts and the triggering of alternative interpretations. As spontaneous perceptual reversals may occur even in the absence of attention to these stimuli, we investigated neural activity patterns in response to perceptual changes of ambiguous Necker cube under different amounts of working memory load using a dual-task design. We hypothesized that the same regions that process working memory load are involved in perceptual switching and confirmed the prediction that perceptual reversals led to fMRI responses that linearly depended on load. Accordingly, posterior Superior Parietal Lobule, anterior Prefrontal and Dorsolateral Prefrontal cortices exhibited differential BOLD signal changes in response to perceptual reversals under working memory load. Our results also suggest that the posterior Superior Parietal Lobule may be directly involved in the emergence of perceptual reversals, given that it specifically reflects both perceptual versus real changes and load levels. The anterior Prefrontal and Dorsolateral Prefrontal cortices, showing a significant interaction between reversal levels and load, might subserve a modulatory role in such reversals, in a mirror symmetric way: in the former activation is suppressed by the highest loads, and in the latter deactivation is reduced by highest loads, suggesting a more direct role of the aPFC in reversal generation.

Evidence for distinct levels of neural adaptation to both coherent and incoherently moving visual surfaces in visual area hMT

&NA; Visual adaptation describes the processes by which the visual system alters its operating properties in response to changes in the environment. It is one of the mechanisms controlling visual perceptual bistability – when two perceptual solutions are available – by controlling the duration of each percept. Moving plaids are an example of such ambiguity. They can be perceived as two surfaces sliding incoherently over each other or as a single coherent surface. Here, we investigated, using fMRI, whether activity in the human motion complex (hMT+), a region tightly related to the perceptual integration of visual motion, is modulated by distinct forms of visual adaptation to coherent or incoherent perception of moving plaids. Our hypothesis is that exposure to global coherent or incoherent moving stimuli leads to different levels of measurable adaptation, reflected in hMT+ activity. We found that the strength of the measured visual adaptation effect depended on whether subjects integrated (coherent percept) or segregated (incoherent percept) surface motion signals. Visual motion adaptation was significant both for coherent motion and globally incoherent surface motion. Although not as strong as to the coherent percept, visual adaptation due to the incoherent percept also affects hMT+. This shows that adaptation can contribute to regulate percept duration during visual bistability, with distinct weights, depending on the type of percept. Our findings suggest a link between bistability and adaptation mechanisms, both due to coherent and incoherent motion percepts, but in an asymmetric manner. These asymmetric adaptation weights have strong implications in models of perceptual decision and may explain asymmetry of perceptual interpretation periods. HighlightsPerceptual adaptation occurs for both coherent and incoherent moving surfaces.Adaptation mechanisms due to coherent and incoherent moving surfaces are asymmetric.Asymmetric adaptation weights may explain imbalances in bistable perception.

Extending Inferential Group Analysis in Type 2 Diabetic Patients with Multivariate GLM Implemented in SPM8

Background: Although voxel based morphometry studies are still the standard for analyzing brain structure, their dependence on massive univariate inferential methods is a limiting factor. A better understanding of brain pathologies can be achieved by applying inferential multivariate methods, which allow the study of multiple dependent variables, e.g. different imaging modalities of the same subject. Objective: Given the widespread use of SPM software in the brain imaging community, the main aim of this work is the implementation of massive multivariate inferential analysis as a toolbox in this software package. applied to the use of T1 and T2 structural data from diabetic patients and controls. This implementation was compared with the traditional ANCOVA in SPM and a similar multivariate GLM toolbox (MRM). Method: We implemented the new toolbox and tested it by investigating brain alterations on a cohort of twenty-eight type 2 diabetes patients and twenty-six matched healthy controls, using information from both T1 and T2 weighted structural MRI scans, both separately – using standard univariate VBM - and simultaneously, with multivariate analyses. Results: Univariate VBM replicated predominantly bilateral changes in basal ganglia and insular regions in type 2 diabetes patients. On the other hand, multivariate analyses replicated key findings of univariate results, while also revealing the thalami as additional foci of pathology. Conclusion: While the presented algorithm must be further optimized, the proposed toolbox is the first implementation of multivariate statistics in SPM8 as a user-friendly toolbox, which shows great potential and is ready to be validated in other clinical cohorts and modalities.

Permutations of functional magnetic resonance imaging classification may not be normally distributed

A fundamental question that often occurs in statistical tests is the normality of distributions. Countless distributions exist in science and life, but one distribution that is obtained via permutations, usually referred to as permutation distribution, is interesting. Although a permutation distribution should behave in accord with the central limit theorem, if both the independence condition and the identical distribution condition are fulfilled, no studies have corroborated this concurrence in functional magnetic resonance imaging data. In this work, we used Anderson–Darling test to evaluate the accordance level of permutation distributions of classification accuracies to normality expected under central limit theorem. A simulation study has been carried out using functional magnetic resonance imaging data collected, while human subjects responded to visual stimulation paradigms. Two scrambling schemes are evaluated: the first based on permuting both the training and the testing sets and the second on permuting only the testing set. The results showed that, while a normal distribution does not adequately fit to permutation distributions most of the times, it tends to be quite well acceptable when mean classification accuracies averaged over a set of different classifiers is considered. The results also showed that permutation distributions can be probabilistically affected by performing motion correction to functional magnetic resonance imaging data, and thus may weaken the approximation of permutation distributions to a normal law. Such findings, however, have no relation to univariate/univoxel analysis of functional magnetic resonance imaging data. Overall, the results revealed a strong dependence across the folds of cross-validation and across functional magnetic resonance imaging runs and that may hinder the reliability of using cross-validation. The obtained p-values and the drawn confidence level intervals exhibited beyond doubt that different permutation schemes may beget different permutation distributions as well as different levels of accord with central limit theorem. We also found that different permutation schemes can lead to different permutation distributions and that may lead to different assessment of the statistical significance of classification accuracy.