A deep learning based approach identifies regions more relevant than resting-state networks to the prediction of general intelligence from resting-state fMRI.

Prediction of cognitive ability latent factors such as general intelligence from neuroimaging has elucidated questions pertaining to their neural origins. However, predicting general intelligence from functional connectivity limit hypotheses to that specific domain, being agnostic to time-distributed features and dynamics. We used an ensemble of recurrent neural networks to circumvent this limitation, bypassing feature extraction, to predict general intelligence from resting-state functional magnetic resonance imaging regional signals of a large sample (n = 873) of Human Connectome Project adult subjects. Ablating common resting-state networks (RSNs) and measuring degradation in performance, we show that model reliance can be mostly explained by network size. Using our approach based on the temporal variance of saliencies, that is, gradients of outputs with regards to inputs, we identify a candidate set of networks that more reliably affect performance in the prediction of general intelligence than similarly sized RSNs. Our approach allows us to further test the effect of local alterations on data and the expected changes in derived metrics such as functional connectivity and instantaneous innovations.

Evidence of regional associations between age-related inter-individual differences in resting-state functional connectivity and cortical thinning revealed through a multi-level analysis.

Normal aging incurs functional and anatomical alterations in the brain. Cortical thinning, age-related alterations in resting-state functional connectivity (RSFC) and reductions in fractional amplitude of low frequency fluctuations (fALFF) are key components of brain aging that can be studied by neuroimaging. However, the level of association between these processes has not been fully established. We performed an analysis at multiple-levels, i.e. region or connection and modality, to investigate whether the evidence for the effect of aging on fALFF, RSFC and cortical thickness are associated in a large cohort. Our results show that there is a positive association between the level of evidence of age-related effects in all three in the brain. We also demonstrate that on a regional basis the association between RSFC alterations and cortical atrophy may be either positive or negative, which may relate to compensatory mechanisms predicted by the Scaffolding Theory of Aging and Cognition (STAC).

A dual 1H/19F birdcage coil for small animals at 7 T MRI.

OBJECTIVE: Given the growing interest in fluorine, it is necessary to develop new multi-tuned RF coils. Therefore, our objective is to design a simple and versatile double-tuned RF coil that can be used as a transmitter and receiver double-tuned coil (1H and 19F) or as transmitter-only double-tuned coil. MATERIALS AND METHODS: A high-pass eight-element birdcage coil was built for 1H and 19F for a 7 T scanner. PIN diodes and cable traps to block unwanted common mode currents in cables were introduced to confer more flexibility to the coil. S-parameters and quality factor were measured in workbench and signal to noise ratio as well as signal intensity profiles in imaging experiments. RESULTS: Bench measurements show S11 values less than - 33 dB, S21 lower than - 13 dB and quality factors ratio of the order of 1.8 that are in agreement with good performances of a RF coil, as well as values of - 39 dB for 19F and - 30 dB for 1H as good detuning values. Signal intensity profiles prove excellent homogeneity at 1H and 19F. DISCUSSION: We present a simple structure of a double-tuned high-pass birdcage coil tuned to 1H and 19F that shows a great uniformity and sensitivity for 19F.

A principled multivariate intersubject analysis of generalized partial directed coherence with Dirichlet regression: Application to healthy aging in areas exhibiting cortical thinning.

Background Generalized Partial Directed Coherence (GPDC) is a multivariate measure of predictability between functional timeseries defined in the frequency domain. However, analysis has often been constrained by its compositional nature. Specifically, the squared GPDC from a node region to all nodes in any given frequency must sum to one. New method When analyzing GPDC spectra, it is imperative to consider that squared GPDC from a source timeseries sums to one over its target timeseries. Dirichlet Regression allows the modeling of compositional data and, therefore, becomes a principled choice for the multivariate analysis of GPDC on arbitrary subject-level variables. Results Eleven resting-state fMRI connections underwent age-related alterations, with two decreases in squared GPDC from a region to itself in two frequencies, signaling increased integration with the rest, and nine increases in squared GPDC, one involving different regions. All frequencies had at least one alteration due to age. Comparison with existing method(s) Our methodology identifies alterations in GPDC in more connections than a naïve approach based on linear regression and centered log-ratio analysis. We also studied alternative connectivity indices between the same ROIs, uncovering no effect of age on the time-domain predictive-causality metrics for any connection, while for Pearson correlation five connections displayed significant effects of age, with parallels to the results pertaining to GPDC. Conclusions Dirichlet Regression allows the study of continuous or discrete variables as predictors for the analysis of GPDC, enabling a wider adoption of this measure of connectivity.

Neuro-degeneration profile of Alzheimer's patients: A brain morphometry study.

INTRODUCTION: Alzheimer's disease (AD) is a primary and progressive neurodegenerative disorder, which is marked by cognitive deterioration and memory impairment. Atrophy of hippocampus and other basal brain regions is one of the most predominant structural imaging findings related to AD. Most studies have evaluated the pre-clinical and initial stages of AD through clinical trials using Magnetic Resonance Imaging. Structural biomarkers for advanced AD stages have not been evaluated yet, being considered only hypothetically. OBJECTIVE: To evaluate the brain morphometry of AD patients at all disease stages, identifying the structural neuro-degeneration profile associated with AD severity. MATERIAL AND METHODS: AD patients aged 60 years or over at different AD stages were recruited and grouped into three groups following the Clinical Dementia Rating (CDR) score: CDR1 (n = 16), CDR2 (n = 15), CDR3 (n = 13). Age paired healthy volunteers (n = 16) were also recruited (control group). Brain images were acquired on a 3T magnetic resonance scanner using a conventional Gradient eco 3D T1-w sequence without contrast injection. Volumetric quantitative data and cortical thickness were obtained by automatic segmentation using the Freesurfer software. Volume of each brain region was normalized by the whole brain volume in order to minimize age and body size effects. Volume and cortical thickness variations among groups were compared. RESULTS: Atrophy was observed in the hippocampus, amygdala, entorhinal cortex, parahippocampal region, temporal pole and temporal lobe of patients suffering from AD at any stage. Cortical thickness was reduced only in the parahippocampal gyrus at all disease stages. Volume and cortical thickness were correlated with the Mini Mental State Examination (MMSE) score in all studied regions, as well as with CDR and disease duration. DISCUSSION AND CONCLUSION: As previously reported, brain regions affected by AD during its initial stages, such as hippocampus, amygdala, entorhinal cortex, and parahippocampal region, were found to be altered even in individuals with severe AD. In addition, individuals, specifically, with CDR 3, have multiple regions with lower volumes than individuals with a CDR 2. These results indicate that rates of atrophy have not plateaued out at CDR 2-3, and in severe patients there are yet neuronal loss and gliosis. These findings can add important information to the more accepted model in the literature that focuses mainly on early stages. Our findings allow a better understanding on the AD pathophysiologic process and follow-up process of drug treatment even at advanced disease stages.

Effects of reduced oxygen availability on the vascular response and oxygen consumption of the activated human visual cortex.

PURPOSE: To identify the impact of reduced oxygen availability on the evoked vascular response upon visual stimulation in the healthy human brain by magnetic resonance imaging (MRI). MATERIALS AND METHODS: Functional MRI techniques based on arterial spin labeling (ASL), blood oxygenation level-dependent (BOLD), and vascular space occupancy (VASO)-dependent contrasts were utilized to quantify the BOLD signal, cerebral blood flow (CBF), and volume (CBV) from nine subjects at 3T (7M/2F, 27.3 ± 3.6 years old) during normoxia and mild hypoxia. Changes in visual stimulus-induced oxygen consumption rates were also estimated with mathematical modeling. RESULTS: Significant reductions in the extension of activated areas during mild hypoxia were observed in all three imaging contrasts: by 42.7 ± 25.2% for BOLD (n = 9, P = 0.002), 33.1 ± 24.0% for ASL (n = 9, P = 0.01), and 31.9 ± 15.6% for VASO images (n = 7, P = 0.02). Activated areas during mild hypoxia showed responses with similar amplitude for CBF (58.4 ± 18.7% hypoxia vs. 61.7 ± 16.1% normoxia, P = 0.61) and CBV (33.5 ± 17.5% vs. 25.2 ± 13.0%, P = 0.27), but not for BOLD (2.5 ± 0.8% vs. 4.1 ± 0.6%, P = 0.009). The estimated stimulus-induced increases of oxygen consumption were smaller during mild hypoxia as compared to normoxia (3.1 ± 5.0% vs. 15.5 ± 15.1%, P = 0.04). CONCLUSION: Our results demonstrate an altered vascular and metabolic response during mild hypoxia upon visual stimulation. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:142-149.

Pre- and postshunting magnetization transfer ratios are in accordance with neurological and behavioral changes in hydrocephalic immature rats.

Hydrocephalus is a common neurological condition in children characterized by an imbalance between the production and absorption of cerebrospinal fluid (CSF), causing abnormal fluid accumulation in the brain cavities. Shunt systems have been used to drain excess CSF and to prevent progressive ventricular enlargement. However, despite improvements in these systems, neurological and structural changes cannot always be reversed. Our aim was to evaluate the magnetization transfer ratio as a biomarker for the effectiveness of a CSF shunt system to treat neurological and behavioral disorders observed in experimental hydrocephalus. Seven-day-old Wistar rats were used in this study. The pups were subjected to hydrocephalus induction via 20% kaolin intracisternal injection. After confirmation of ventriculomegaly by magnetic resonance imaging (MRI), a group of animals underwent placement of a ventriculosubcutaneous shunt (VSS). The reduction in ventricular size in hydrocephalic rats operated with functional VSS was observed as a decrease in ventricular ratio values and preservation of the corpus callosum thickness. Magnetization transfer values were significantly increased and matched to the recovery process of axonal myelination observed based on more-intense blue staining by solochrome cyanin. The histopathological analysis revealed a reduction in reactive astrocytes by means of GFAP immunostaining. The hydrocephalic rats operated with functional VSS also showed significant progress in motor and exploratory activities, similar to the control animals, at the end of the experiment. In conclusion, the VSS system employed 7 days after hydrocephalus induction was able to prevent structural damage and restore the axonal myelination process in periventricular structures by stabilizing and reducing the ventricular enlargement, and the results are in accordance with the magnetization transfer ratio in MRI.