INTRODUCTION:Effective longitudinal biomarkers that track disease progression are needed to characterize the presymptomatic phase of genetic frontotemporal dementia (FTD). We investigate the utility of cerebral perfusion as one such biomarker in presymptomatic FTD mutation carriers. METHODS:We investigated longitudinal profiles of cerebral perfusion using arterial spin labeling magnetic resonance imaging in 42 C9orf72, 70 GRN, and 31 MAPT presymptomatic carriers and 158 non-carrier controls. Linear mixed effects models assessed perfusion up to 5 years after baseline assessment. RESULTS:Perfusion decline was evident in all three presymptomatic groups in global gray matter. Each group also featured its own regional pattern of hypoperfusion over time, with the left thalamus common to all groups. Frontal lobe regions featured lower perfusion in those who symptomatically converted versus asymptomatic carriers past their expected age of disease onset. DISCUSSION:Cerebral perfusion is a potential biomarker for assessing genetic FTD and its genetic subgroups prior to symptom onset. HIGHLIGHTS:Gray matter perfusion declines in at-risk genetic frontotemporal dementia (FTD). Regional perfusion decline differs between at-risk genetic FTD subgroups . Hypoperfusion in the left thalamus is common across all presymptomatic groups. Converters exhibit greater right frontal hypoperfusion than non-converters past their expected conversion date. Cerebral hypoperfusion is a potential early biomarker of genetic FTD.

BackgroundAbnormalities in cerebral blood flow (CBF) are common in bipolar disorder (BD). Despite known differences in CBF between healthy adolescent males and females, sex differences in CBF among adolescents with BD have never been studied. ObjectiveTo examine sex differences in CBF among adolescents with BD versus healthy controls (HC). MethodsCBF images were acquired using arterial spin labeling (ASL) perfusion magnetic resonance imaging (MRI) in 123 adolescents (72 BD: 30M, 42F; 51 HC: 22M, 29F) matched for age (13-20 years). Whole brain voxel-wise analysis was performed in a general linear model with sex and diagnosis as fixed factors, sex-diagnosis interaction effect, and age as a covariate. We tested for main effects of sex, diagnosis, and their interaction. Results were thresholded at cluster forming p = 0.0125, with posthoc Bonferroni correction (p = 0.05/4 groups). ResultsA main effect of diagnosis (BD > HC) was observed in the superior longitudinal fasciculus (SLF), underlying the left precentral gyrus (F =10.24 (3), p < 0.0001). A main effect of sex (F > M) on CBF was detected in the precuneus/posterior cingulate cortex (PCC), left frontal and occipital poles, left thalamus, left SLF, and right inferior longitudinal fasciculus (ILF). No regions demonstrated a significant sex-by-diagnosis interaction. Exploratory pairwise testing in regions with a main effect of sex revealed greater CBF in females with BD versus HC in the precuneus/PCC (F = 7.1 (3), p < 0.01). ConclusionGreater CBF in female adolescents with BD versus HC in the precuneus/PCC may reflect the role of this region in the neurobiological sex differences of adolescent-onset BD. Larger studies targeting underlying mechanisms, such as mitochondrial dysfunction or oxidative stress, are warranted.

Anatomical T1-weighted imaging allows us to examine relationships of regional morphology across the brain through structural covariance. Here, we investigated structural covariance stability using decreasing amounts of T1-weighted imaging data from the UK Biobank. Starting from 1,753 individuals, we found that it is possible to drastically reduce the sample and still maintain adequate stability (78% agreement with ~87 individuals). We note, however, that stability was regionally variable; lateral and cortical regions were least affected by sample size while medial and subcortical regions were most affected. These findings may inform sample size considerations for MRI-based structural covariance in large population studies.

PurposeCardiac-related intracranial pulsatility may relate to cerebrovascular health, and this information is contained in BOLD MRI data. There is broad interest in methods to isolate BOLD pulsatility, and the current study examines a deep learning approach. MethodsMulti-echo BOLD images, respiratory, and cardiac recordings were measured in 55 adults. Ground truth BOLD pulsatility maps were calculated with an established method. BOLD fast Fourier transform magnitude images were used as temporal-frequency image inputs to a U-Net deep learning model. Model performance was evaluated by mean squared error (MSE), mean absolute error (MAE), structural similarity index (SSIM), and mutual information (MI). Experiments evaluated the influence of input channel size, an age group effect during training, dependence on TE, performance without the U-Net architecture, and importance of respiratory preprocessing. ResultsThe U-Net model generated BOLD pulsatility maps with lower MSE as additional fast Fourier transform input images were used. There was no age group effect for MSE (P > 0.14). MAE and SSIM metrics did not vary across TE (P > 0.36), whereas MI showed a significant TE dependence (P < 0.05). The U-Net versus no U-Net comparison showed no significant difference for MAE (P = 0.059); however, SSIM and MI were significantly different between models (P < 0.001). Within the insula, the cross-correlation values were high (r > 0.90) when comparing the U-Net model trained with/without respiratory preprocessing. ConclusionMulti-echo BOLD pulsatility maps were synthesized from a U-net model that was trained to use temporal-frequency BOLD image inputs. This work adds to the deep learning methods that characterize BOLD physiological signals.

BACKGROUND:Clinical neuroimaging studies often investigate group differences between patients and controls, yet multivariate imaging features may enable individual-level classification. This study aims to classify youth with bipolar disorder (BD) versus healthy youth using grey matter cerebral blood flow (CBF) data analyzed with logistic regressions. METHODS:Using a 3 Tesla magnetic resonance imaging (MRI) system, we collected pseudo-continuous, arterial spin-labelling, resting-state functional MRI (rfMRI) and T 1-weighted images from youth with BD and healthy controls. We used 3 logistic regression models to classify youth with BD versus controls, controlling for age and sex, using mean grey matter CBF as a single explanatory variable, quantitative CBF features based on principal component analysis (PCA) or relative (intensity-normalized) CBF features based on PCA. We also carried out a comparison analysis using rfMRI data. RESULTS:The study included 46 patients with BD (mean age 17 yr, standard deviation [SD] 1 yr; 25 females) and 49 healthy controls (mean age 16 yr, SD 2 yr; 24 females). Global mean CBF and multivariate quantitative CBF offered similar classification performance that was above chance. The association between CBF images and the feature map was not significantly different between groups (p = 0.13); however, the multivariate classifier identified regions with lower CBF among patients with BD (ΔCBF = -2.94 mL/100 g/min; permutation test p = 0047). Classification performance decreased when considering rfMRI data. LIMITATIONS:We cannot comment on which CBF principal component is most relevant to the classification. Participants may have had various mood states, comorbidities, demographics and medication records. CONCLUSION:Brain CBF features can classify youth with BD versus healthy controls with above-chance accuracy using logistic regression. A global CBF feature may offer similar classification performance to distinct multivariate CBF features.

White matter hyperintensities (WMHs) are emblematic of cerebral small vessel disease, yet effects on the brain have not been well characterized at midlife. Here, we investigated whether WMH volume is associated with brain network alterations in midlife adults. Two hundred and fifty-four participants from the Coronary Artery Risk Development in Young Adults study were selected and stratified by WMH burden into Lo-WMH (mean age = 50 +/- 3.5 years) and Hi-WMH (mean age = 51 +/- 3.7 years) groups of equal size. We constructed group-level covariance networks based on cerebral blood flow (CBF) and gray matter volume (GMV) maps across 74 gray matter regions. Through consensus clustering, we found that both CBF and GMV covariance networks partitioned into modules that were largely consistent between groups. Next, CBF and GMV covariance network topologies were compared between Lo- and Hi-WMH groups at global (clustering coefficient, characteristic path length, global efficiency) and regional (degree, betweenness centrality, local efficiency) levels. At the global level, there were no between-group differences in either CBF or GMV covariance networks. In contrast, we found between-group differences in the regional degree, betweenness centrality, and local efficiency of several brain regions in both CBF and GMV covariance networks. Overall, CBF and GMV covariance analyses provide evidence that WMH-related network alterations are present at midlife.

Purpose Develop and evaluate a deep learning approach to estimate cerebral blood flow (CBF) and arterial transit time (ATT) from multiple post-labeling delay (PLD) ASL MRI. Methods ASL MRI were acquired with 6 PLDs on a 1.5T or 3T GE system in adults with and without cognitive impairment (N = 99). Voxel-level CBF and ATT maps were quantified by training models with distinct convolutional neural network architectures: (1) convolutional neural network (CNN) and (2) U-Net. Models were trained and compared via 5-fold cross validation. Performance was evaluated using mean absolute error (MAE). Model outputs were trained on and compared against a reference ASL model fitting after data cleaning. Minimally processed ASL data served as another benchmark. Model output uncertainty was estimated using Monte Carlo dropout. The better-performing neural network was subsequently re-trained on inputs with missing PLDs to investigate generalizability to different PLD schedules. Results Relative to the CNN, the U-Net yielded lower MAE on training data. On test data, the U-Net MAE was 8.4 +/- 1.4 mL/100 g/min for CBF and 0.22 +/- 0.09 s for ATT. A significant association was observed between MAE and Monte Carlo dropout-based uncertainty estimates. Neural network performance remained stable despite systematically reducing the number of input images (i.e., up to 3 missing PLD images). Mean processing time was 10.77 s for the U-Net neural network compared to 10 min 41 s for the reference pipeline. Conclusion It is feasible to generate CBF and ATT maps from 1.5T and 3T multi-PLD ASL MRI with a fast deep learning image-generation approach.

Background Bipolar disorder (BD) is associated with elevated body mass index (BMI) and increased rates of obesity. Obesity among individuals with BD is associated with more severe course of illness. Motivated by previous research on BD and BMI in youth as well as brain findings in the reward circuit, the current study investigates differences in cerebral blood flow (CBF) in youth BD with and without comorbid overweight/obesity (OW/OB). Methods Participants consisted of youth, ages 13-20 years, including BD with OW/OB (BDOW/OB; n = 25), BD with normal weight (BDNW; n = 55), and normal-weight healthy controls (HC; n = 61). High-resolution T1-weighted and pseudo-continuous arterial spin labeling images were acquired using 3 Tesla magnetic resonance imaging. CBF differences were assessed using both region of interest and whole-brain voxel-wise approaches. Results Voxel-wise analysis revealed significantly higher CBF in reward-associated regions in the BDNW group relative to the HC and BDOW/OB groups. CBF did not differ between the HC and BDOW/OB groups. There were no significant region of interest findings. Conclusions The current study identified distinct CBF levels relating to BMI in BD in the reward circuit, which may relate to underlying differences in cerebral metabolism, compensatory effects, and/or BD severity. Future neuroimaging studies are warranted to examine for changes in the CBF-OW/OB link over time and in relation to treatment.

White matter hyperintensities (WMHs) are emblematic of cerebral small vessel disease, yet characterization at midlife is poorly studied. Here, we investigated whether WMH volume is associated with brain network alterations in midlife adults. 254 participants from the Coronary Artery Risk Development in Young Adults (CARDIA) study were selected and stratified by WMH burden yielding two groups of equal size (Lo- and Hi-WMH groups). We constructed group-level covariance networks based on cerebral blood flow (CBF) and grey matter volume (GMV) maps across 74 grey matter regions. Through consensus clustering, we found that both CBF and GMV covariance networks were partitioned into modules that were largely consistent between groups. Next, CBF and GMV covariance network topologies were compared between Lo- and Hi-WMH groups at global (clustering coefficient, characteristic path length, global efficiency) and regional (degree, betweenness centrality, local efficiency) levels. At the global level, there were no group differences in either CBF or GMV covariance networks. In contrast, we found group differences in the regional degree, betweenness centrality, and local efficiency of several brain regions in both CBF and GMV covariance networks. Overall, CBF and GMV covariance analyses provide evidence of WMH-related network alterations that were observed at midlife.

The purpose of this work was to develop and evaluate a deep learning approach for estimation of cerebral blood flow (CBF) and arterial transit time (ATT) from multiple post-label delay (PLD) arterial spin-labelled (ASL) MRI. Six-PLD ASL MRI was acquired on a 1.5T or 3T system among 99 older males and females with and without cognitive impairment. We trained and compared two network architectures: standard feed-forward convolutional neural network (CNN) and U-Net. Mean absolute error (MAE) was evaluated between model estimates and ground truth obtained through conventional processing. The best-performing model was re-trained on inputs with missing PLDs to investigate generalizability to different PLD schedules. Relative to the CNN, the U-Net yielded lower MAE on training data. On test data, the U-Net MAE was 8.4±1.4 ml/100g/min for CBF and 0.22±0.09 s for ATT. Model uncertainty, estimated with Monte Carlo dropout, was associated with model error. Network estimates remained stable when tested on inputs with up to three missing PLD images. Mean processing times were: U-Net pipeline = 10.77s; ground truth pipeline = 10min 41s. These results suggest hemodynamic parameter estimation from 1.5T and 3T multi-PLD ASL MRI is feasible and fast with a deep learning image-generation approach.

Adolescence is a period of rapid development of the brain's inherent functional and structural networks; however, little is known about the region-to-region organization of adolescent cerebral blood flow (CBF) or its relationship to neuroanatomy. Here, we investigate both the regional covariation of CBF MRI and the covariation of structural MRI, in adolescents with and without bipolar disorder. Bipolar disorder is a disease with increased onset during adolescence, putative vascular underpinnings, and evidence of anomalous CBF and brain structure. In both groups, through hierarchical clustering, we found CBF covariance was principally described by clusters of regions circumscribed to the left hemisphere, right hemisphere, and the inferior brain; these clusters were spatially reminiscent of cerebral vascular territories. CBF covariance was associated with structural covariance in both the healthy group (n = 56; r = 0.20, p < 0.0001) and in the bipolar disorder group (n = 68; r = 0.36, p < 0.0001), and this CBF-structure correspondence was higher in bipolar disorder (p = 0.0028). There was lower CBF covariance in bipolar disorder compared to controls between the left angular gyrus and pre- and post-central gyri. Altogether, CBF covariance revealed distinct brain organization, had modest correspondence to structural covariance, and revealed evidence of differences in bipolar disorder.

Neuroimaging reveals differences between Bipolar Disorder (BD) and controls, even at early ages in life. Conventional univariate statistics does not permit examination of between-subject differences. This study investigates differences between adolescents with and without BD through comparison of region-by-region co-variation of Cerebral Blood Flow and Grey Matter density.

We report results from 21-cm intensity maps acquired from the Parkes radio telescope and cross-correlatedwith galaxy maps from the 2dF galaxy survey. The data span the redshift range 0.057 < z < 0.098 and cover approximately 1300 deg(2) over two long fields. Cross-correlation is detected at a significance of 5.7 sigma. The amplitude of the cross-power spectrum is low relative to the expected dark matter power spectrum, assuming a neutral hydrogen (H-I) bias andmass density equal to measurements from the ALFALFA survey. The decrement is pronounced and statistically significant at small scales. At k similar to 1.5 h Mpc(-1), the cross-power spectrum is more than a factor of 6 lower than expected, with a significance of 15.3 sigma. This decrement indicates a lack of clustering of neutral hydrogen (H-I), a small correlation coefficient between optical galaxies and H-I, or some combination of the two. Separating 2dF into red and blue galaxies, we find that red galaxies are much more weakly correlated with H-I on k similar to 1.5 h Mpc(-1) scales, suggesting that H-I is more associated with blue star-forming galaxies and tends to avoid red galaxies.

We report results from 21-cm intensity maps acquired from the Parkes radio telescope and cross-correlated with galaxy maps from the 2dF galaxy survey. The data span the redshift range $0.057u003czu003c0.098$ and cover approximately 1,300 square degrees over two long fields. Cross correlation is detected at a significance of $5.18sigma$. The amplitude of the cross-power spectrum is low relative to the expected dark matter power spectrum, assuming a neutral hydrogen (HI) bias and mass density equal to measurements from the ALFALFA survey. The decrement is pronounced and statistically significant at small scales. At $ksim1.5$ $ h mathrm{Mpc^{-1}}$, the cross power spectrum is more than a factor of 6 lower than expected, with a significance of $14.8,sigma$. This decrement indicates either a lack of clustering of neutral hydrogen (HI), a small correlation coefficient between optical galaxies and HI, or some combination of the two. Separating 2dF into red and blue galaxies, we find that red galaxies are much more weakly correlated with HI on $ksim1.5$ $h mathrm{Mpc^{-1}}$ scales, suggesting that HI is more associated with blue star-forming galaxies and tends to avoid red galaxies.

Measuring intracranial vascular compliance (VC) may help characterize a patient’s resilience to Alzheimer’s disease and dementia. In this study, we estimate VC from a BOLD MRI cerebrovascular reactivity (CVR) protocol by measuring cardiac-related BOLD pulsatility and end-tidal CO 2 during periods of hypercapnia and normocapnia. We measured high VC in regions proximal to major cerebral arteries like the thalamus and insula, and low VC in the parietal lobe. This study demonstrates the feasibility of extracting additional information from BOLD-CVR experiments and suggests regional differences in VC. Future work will include a second cohort to estimate VC in different patient populations.

Persistent exposure to highly pulsatile blood can damage the brain’s microvasculature. A convenient method for measuring cerebral pulsatility would allow investigation into its relationship with vascular dysfunction and cognitive decline. In this work, we propose a convolutional neural network (CNN) based deep learning solution to estimate cerebral pulsatility using only the frequency content from BOLD MRI scans. Various frequency component inputs were assessed, and echo time dependence was evaluated with a 5-fold cross-validation. Pulsatility was estimated from BOLD MRI data acquired on a different scanner to assess generalizability. The CNN reliably estimated pulsatility and was robust to various scan parameters.