Static and dynamic brain morphological changes in isolated REM sleep behavior disorder compared to normal aging.

Objective/background: To assess whether cerebral structural alterations in isolated rapid eye movement sleep behavior disorder (iRBD) are progressive and differ from those of normal aging and whether they are related to clinical symptoms. Patients/methods: In a longitudinal study of 18 patients with iRBD (age, 66.1 ± 5.7 years; 13 males; follow-up, 1.6 ± 0.6 years) and 24 age-matched healthy controls (age, 67.0 ± 4.9 years; 12 males; follow-up, 2.0 ± 0.9 years), all participants underwent multiple extensive clinical examinations, neuropsychological tests, and magnetic resonance imaging at baseline and follow-up. Surface-based cortical reconstruction and automated subcortical structural segmentation were performed on T1-weighted images. We used mixed-effects models to examine the differences between the groups and the differences in anatomical changes over time. Results: None of the patients with iRBD demonstrated phenoconversion during the follow-up. Patients with iRBD had thinner cortices in the frontal, occipital, and temporal regions, and more caudate atrophy, compared to that in controls. In similar regions, group-by-age interaction analysis revealed that patients with iRBD demonstrated significantly slower decreases in cortical thickness and caudate volume with aging than that observed in controls. Patients with iRBD had lower scores on the Korean version of the Mini-Mental Status Examination (p = 0.037) and frontal and executive functions (p = 0.049) at baseline than those in controls; however, no significant group-by-age interaction was identified. Conclusion: Patients with iRBD show brain atrophy in the regions that are overlapped with the areas that have been documented to be affected in early stages of Parkinson's disease. Such atrophy in iRBD may not be progressive but may be slower than that in normal aging. Cognitive impairment in iRBD is not progressive.

Brain age predicted using graph convolutional neural network explains neurodevelopmental trajectory in preterm neonates.

OBJECTIVES: Dramatic brain morphological changes occur throughout the third trimester of gestation. In this study, we investigated whether the predicted brain age (PBA) derived from graph convolutional network (GCN) that accounts for cortical morphometrics in third trimester is associated with postnatal abnormalities and neurodevelopmental outcome. METHODS: In total, 577 T1 MRI scans of preterm neonates from two different datasets were analyzed; the NEOCIVET pipeline generated cortical surfaces and morphological features, which were then fed to the GCN to predict brain age. The brain age index (BAI; PBA minus chronological age) was used to determine the relationships among preterm birth (i.e., birthweight and birth age), perinatal brain injuries, postnatal events/clinical conditions, BAI at postnatal scan, and neurodevelopmental scores at 30 months. RESULTS: Brain morphology and GCN-based age prediction of preterm neonates without brain lesions (mean absolute error [MAE]: 0.96 weeks) outperformed conventional machine learning methods using no topological information. Structural equation models (SEM) showed that BAI mediated the influence of preterm birth and postnatal clinical factors, but not perinatal brain injuries, on neurodevelopmental outcome at 30 months of age. CONCLUSIONS: Brain morphology may be clinically meaningful in measuring brain age, as it relates to postnatal factors, and predicting neurodevelopmental outcome. CLINICAL RELEVANCE STATEMENT: Understanding the neurodevelopmental trajectory of preterm neonates through the prediction of brain age using a graph convolutional neural network may allow for earlier detection of potential developmental abnormalities and improved interventions, consequently enhancing the prognosis and quality of life in this vulnerable population. KEY POINTS: •Brain age in preterm neonates predicted using a graph convolutional network with brain morphological changes mediates the pre-scan risk factors and post-scan neurodevelopmental outcomes. •Predicted brain age oriented from conventional deep learning approaches, which indicates the neurodevelopmental status in neonates, shows a lack of sensitivity to perinatal risk factors and predicting neurodevelopmental outcomes. •The new brain age index based on brain morphology and graph convolutional network enhances the accuracy and clinical interpretation of predicted brain age for neonates.

White matter microstructure and connectivity changes after surgery in male adults with obstructive sleep apnea: recovery or reorganization?

Study objectives: Obstructive sleep apnea (OSA) is a prevalent clinical problem significantly affecting cognitive functions. Surgical treatment is recommended for those unable to use continuous positive airway pressure. We aimed to investigate the therapeutic effect of upper airway surgery on the white matter (WM) microstructure and brain connectivity in patients with OSA. Methods: Twenty-one male patients with moderate-to-severe OSA were recruited for multi-level upper airway surgery. Overnight polysomnography (PSG), neuropsychiatric tests, and brain MRI scans were acquired before and 6.1 ± 0.8 months after surgery. Nineteen male patients with untreated OSA were also included as a reference group. We calculated the longitudinal changes of diffusion tensor imaging (DTI) parameters, including fractional anisotropy (ΔFA) and mean/axial/radial diffusivity (ΔMD/AD/RD). We also assessed changes in network properties based on graph theory. Results: Surgically treated patients showed improvement in PSG parameters and verbal memory after surgery. Globally, ΔFA was significantly higher and ΔRD was lower in the surgery group than in the untreated group. Especially ΔFA of the tracts involved in the limbic system was higher after surgery. In network analysis, higher Δbetweenness and lower Δclustering coefficients were observed in the surgical group than in the untreated group. Finally, the improvement of verbal memory after surgery positively correlated with ΔFA in superior thalamic radiation (p = 0.021), fronto aslant tracts (p = 0.027), and forceps minor tracts (p = 0.032). Conclusion: Surgical treatment of OSA can alleviate alterations in WM integrity and disruptions in local networks, particularly for the tracts involved in the limbic system. These findings may further explain the cognitive improvement observed after the treatment of OSA.

Style transfer generative adversarial networks to harmonize multisite MRI to a single reference image to avoid overcorrection.

Recent work within neuroimaging consortia have aimed to identify reproducible, and often subtle, brain signatures of psychiatric or neurological conditions. To allow for high-powered brain imaging analyses, it is often necessary to pool MR images that were acquired with different protocols across multiple scanners. Current retrospective harmonization techniques have shown promise in removing site-related image variation. However, most statistical approaches may over-correct for technical, scanning-related, variation as they cannot distinguish between confounded image-acquisition based variability and site-related population variability. Such statistical methods often require that datasets contain subjects or patient groups with similar clinical or demographic information to isolate the acquisition-based variability. To overcome this limitation, we consider site-related magnetic resonance (MR) imaging harmonization as a style transfer problem rather than a domain transfer problem. Using a fully unsupervised deep-learning framework based on a generative adversarial network (GAN), we show that MR images can be harmonized by inserting the style information encoded from a single reference image, without knowing their site/scanner labels a priori. We trained our model using data from five large-scale multisite datasets with varied demographics. Results demonstrated that our style-encoding model can harmonize MR images, and match intensity profiles, without relying on traveling subjects. This model also avoids the need to control for clinical, diagnostic, or demographic information. We highlight the effectiveness of our method for clinical research by comparing extracted cortical and subcortical features, brain-age estimates, and case-control effect sizes before and after the harmonization. We showed that our harmonization removed the site-related variances, while preserving the anatomical information and clinical meaningful patterns. We further demonstrated that with a diverse training set, our method successfully harmonized MR images collected from unseen scanners and protocols, suggesting a promising tool for ongoing collaborative studies. Source code is released in USC-IGC/style_transfer_harmonization (github.com).

Brain BOLD and NIRS response to hyperoxic challenge in sickle cell disease and chronic anemias.

PURPOSE: Congenital anemias, including sickle cell anemia and thalassemia, are associated with cerebral tissue hypoxia and heightened stroke risks. Recent works in sickle cell disease mouse models have suggested that hyperoxia respiratory challenges can identify regions of the brain having chronic tissue hypoxia. Therefore, this work investigated differences in hyperoxic response and regional cerebral oxygenation between anemic and healthy subjects. METHODS: A cohort of 38 sickle cell disease subjects (age 22 ± 8 years, female 39%), 25 non-sickle anemic subjects (age 25 ± 11 years, female 52%), and 31 healthy controls (age 25 ± 10 years, female 68%) were examined. A hyperoxic gas challenge was performed with concurrent acquisition of blood oxygen level-dependent (BOLD) MRI and near-infrared spectroscopy (NIRS). In addition to hyperoxia-induced changes in BOLD and NIRS, global measurements of cerebral blood flow, oxygen delivery, and cerebral metabolic rate of oxygen were obtained and compared between the three groups. RESULTS: Regional BOLD changes were not able to identify brain regions of flow limitation in chronically anemic patients. Higher blood oxygen content and tissue oxygenation were observed during hyperoxia gas challenge. Both control and anemic groups demonstrated lower blood flow, oxygen delivery, and metabolic rate compared to baseline, but the oxygen metabolism in anemic subjects were abnormally low during hyperoxic exposure. CONCLUSION: These results indicated that hyperoxic respiratory challenge could not be used to identify chronically ischemic brain. Furthermore, the low hyperoxia-induced metabolic rate suggested potential negative effects of prolonged oxygen therapy and required further studies to evaluate the risk for hyperoxia-induced oxygen toxicity and cerebral dysfunction.

Tract-specific analysis and neurocognitive functioning in sickle cell patients without history of overt stroke.

INTRODUCTION: Sickle cell disease (SCD) is a hereditary blood disorder in which the oxygen-carrying hemoglobin molecule in red blood cells is abnormal. SCD patients are at increased risks for strokes and neurocognitive deficit, even though neurovascular screening and treatments have lowered the rate of overt strokes. Tract-specific analysis (TSA) is a statistical method to evaluate microstructural WM damage in neurodegenerative disorders, using diffusion tensor imaging (DTI). METHODS: We utilized TSA and compared 11 major brain WM tracts between SCD patients with no history of overt stroke, anemic controls, and healthy controls. We additionally examined the relationship between the most commonly used DTI metric of WM tracts and neurocognitive performance in the SCD patients and healthy controls. RESULTS: Disruption of WM microstructure orientation-dependent metrics for the SCD patients was found in the genu of the corpus callosum (CC), cortico-spinal tract, inferior fronto-occipital fasciculus, right inferior longitudinal fasciculus, superior longitudinal fasciculus, and left uncinate fasciculus. Neurocognitive performance indicated slower processing speed and lower response inhibition skills in SCD patients compared to controls. TSA abnormalities in the CC were significantly associated with measures of processing speed, working memory, and executive functions. CONCLUSION: Decreased DTI-derived metrics were observed on six tracts in chronically anemic patients, regardless of anemia subtype, while two tracks with decreased measures were unique to SCD patients. Patients with WMHs had more significant FA abnormalities. Decreased FA values in the CC significantly correlated with all nine neurocognitive tests, suggesting a critical importance for CC in core neurocognitive processes.

Style Transfer Using Generative Adversarial Networks for Multi-Site MRI Harmonization.

Large data initiatives and high-powered brain imaging analyses require the pooling of MR images acquired across multiple scanners, often using different protocols. Prospective cross-site harmonization often involves the use of a phantom or traveling subjects. However, as more datasets are becoming publicly available, there is a growing need for retrospective harmonization, pooling data from sites not originally coordinated together. Several retrospective harmonization techniques have shown promise in removing cross-site image variation. However, most unsupervised methods cannot distinguish between image-acquisition based variability and cross-site population variability, so they require that datasets contain subjects or patient groups with similar clinical or demographic information. To overcome this limitation, we consider cross-site MRI image harmonization as a style transfer problem rather than a domain transfer problem. Using a fully unsupervised deep-learning framework based on a generative adversarial network (GAN), we show that MR images can be harmonized by inserting the style information encoded from a reference image directly, without knowing their site/scanner labels a priori. We trained our model using data from five large-scale multi-site datasets with varied demographics. Results demonstrated that our style-encoding model can harmonize MR images, and match intensity profiles, successfully, without relying on traveling subjects. This model also avoids the need to control for clinical, diagnostic, or demographic information. Moreover, we further demonstrated that if we included diverse enough images into the training set, our method successfully harmonized MR images collected from unseen scanners and protocols, suggesting a promising novel tool for ongoing collaborative studies.

LEARNING TO SYNTHESIZE CORTICAL MORPHOLOGICAL CHANGES USING GRAPH CONDITIONAL VARIATIONAL AUTOENCODER.

Changes in brain morphology, such as cortical thinning are of great value for understanding the trajectory of brain aging and various neurodegenerative diseases. In this work, we employed a generative neural network variational autoencoder (VAE) that is conditional on age and is able to generate cortical thickness maps at various ages given an input cortical thickness map. To take into account the mesh topology in the model, we proposed a loss function based on weighted adjacency to integrate the surface topography defined as edge connections with the cortical thickness mapped as vertices. Compared to traditional conditional VAE that did not use the surface topological information, our method better predicted "future" cortical thickness maps, especially when the age gap became wider. Our model has the potential to predict the distinctive temporospatial pattern of individual cortical morphology in relation to aging and neurodegenerative diseases.

Quantitative perfusion mapping with induced transient hypoxia using BOLD MRI.

PURPOSE: Gadolinium-based dynamic susceptibility contrast (DSC) is commonly used to characterize blood flow in patients with stroke and brain tumors. Unfortunately, gadolinium contrast administration has been associated with adverse reactions and long-term accumulation in tissues. In this work, we propose an alternative deoxygenation-based DSC (dDSC) method that uses a transient hypoxia gas paradigm to deliver a bolus of paramagnetic deoxygenated hemoglobin to the cerebral vasculature for perfusion imaging. METHODS: Through traditional DSC tracer kinetic modeling, the MR signal change induced by this hypoxic bolus can be used to generate regional perfusion maps of cerebral blood flow, cerebral blood volume, and mean transit time. This gas paradigm and blood-oxygen-level-dependent (BOLD)-MRI were performed concurrently on a cohort of 66 healthy and chronically anemic subjects (age 23.5 ± 9.7, female 64%). RESULTS: Our results showed reasonable global and regional agreement between dDSC and other flow techniques, such as phase contrast and arterial spin labeling. CONCLUSION: In this proof-of-concept study, we demonstrated the feasibility of using transient hypoxia to generate a contrast bolus that mimics the effect of gadolinium and yields reasonable perfusion estimates. Looking forward, optimization of the hypoxia boluses and measurement of the arterial-input function is necessary to improve the accuracy of dDSC. Additionally, a cross-validation study of dDSC and DSC in brain tumor and ischemic stroke subjects is warranted to evaluate the clinical diagnostic utility of this approach.

Transient Hypoxia Model Revealed Cerebrovascular Impairment in Anemia Using BOLD MRI and Near-Infrared Spectroscopy.

BACKGROUND: Obstructive sleep apnea and nocturnal oxygen desaturations, which are prevalent in sickle cell disease (SCD) and chronic anemia disorders, have been linked to risks of stroke and silent cerebral infarcts (SCI). Cerebrovascular response to intermittent desaturations has not been well studied and may identify patients at greatest risk. PURPOSE: To investigate the cerebral dynamic response to induced desaturation in SCD patients with and without SCI, chronic anemia, and healthy subjects. STUDY TYPE: Prospective. SUBJECTS: Twenty-six SCD patients (age = 21 ± 8.2, female 46.2%), including 15 subjects without SCI and nine subjects with SCI, 15 nonsickle anemic patients (age = 22 ± 5.8, female 66.7%), and 31 controls (age = 28 ± 12.3, female 77.4%). FIELD STRENGTH/SEQUENCE: 3T, gradient-echo echo-planar imaging. ASSESSMENT: A transient hypoxia challenge of five breaths of 100% nitrogen gas was performed with blood oxygen level-dependent (BOLD) MRI and near-infrared spectroscopy (NIRS) acquisitions. Hypoxia responses were characterized by desaturation depth, time-to-peak, return-to-baseline half-life, and posthypoxia recovery in the BOLD and NIRS time courses. SCI were documented by T2 fluid-attenuation inversion recovery (FLAIR). STATISTICAL TESTS: Univariate and multivariate regressions were performed between hypoxic parameters and anemia predictors. Voxelwise two-sample t-statistic maps were used to assess the regional difference in hypoxic responses between anemic and control groups. RESULTS: Compared to controls, SCD and chronically anemic patients demonstrated significantly higher desaturation depth (P < 0.01) and shorter return-to-baseline timing response (P < 0.01). Patients having SCI had shorter time-to-peak (P < 0.01), return-to-baseline (P < 0.01), and larger desaturation depth (P < 0.01) in both white matter regions at risk and normal-appearing white matter than patients without infarcts. On multivariate analysis, desaturation depth and timing varied with age, sex, blood flow, white blood cells, and cell-free hemoglobin (r2 = 0.25 for desaturation depth; r2 = 0.18 for time-to-peak; r2 = 0.37 for return-to-baseline). DATA CONCLUSION: Transient hypoxia revealed global and regional response differences between anemic and healthy subjects. SCI was associated with extensive heterogeneity of desaturation dynamics, consistent with extensive underlying microvascular remodeling.

MRI restoration using edge-guided adversarial learning.

Magnetic resonance imaging (MRI) images acquired as multislice two-dimensional (2D) images present challenges when reformatted in orthogonal planes due to sparser sampling in the through-plane direction. Restoring the "missing" through-plane slices, or regions of an MRI image damaged by acquisition artifacts can be modeled as an image imputation task. In this work, we consider the damaged image data or missing through-plane slices as image masks and proposed an edge-guided generative adversarial network to restore brain MRI images. Inspired by the procedure of image inpainting, our proposed method decouples image repair into two stages: edge connection and contrast completion, both of which used general adversarial networks (GAN). We trained and tested on a dataset from the Human Connectome Project to test the application of our method for thick slice imputation, while we tested the artifact correction on clinical data and simulated datasets. Our Edge-Guided GAN had superior PSNR, SSIM, conspicuity and signal texture compared to traditional imputation tools, the Context Encoder and the Densely Connected Super Resolution Network with GAN (DCSRN-GAN). The proposed network may improve utilization of clinical 2D scans for 3D atlas generation and big-data comparative studies of brain morphometry.

White matter has impaired resting oxygen delivery in sickle cell patients.

Although modern medical management has lowered overt stroke occurrence in patients with sickle cell disease (SCD), progressive white matter (WM) damage remains common. It is known that cerebral blood flow (CBF) increases to compensate for anemia, but sufficiency of cerebral oxygen delivery, especially in the WM, has not been systematically investigated. Cerebral perfusion was measured by arterial spin labeling in 32 SCD patients (age range: 10-42 years old, 14 males, 7 with HbSC, 25 HbSS) and 25 age and race-matched healthy controls (age range: 15-45 years old, 10 males, 12 with HbAS, 13 HbAA); 8/24 SCD patients were receiving regular blood transfusions and 14/24 non-transfused SCD patients were taking hydroxyurea. Imaging data from control subjects were used to calculate maps for CBF and oxygen delivery in SCD patients and their T-score maps. Whole brain CBF was increased in SCD patients with a mean T-score of 0.5 and correlated with lactate dehydrogenase (r2 = 0.58, P < 0.0001). When corrected for oxygen content and arterial saturation, whole brain and gray matter (GM) oxygen delivery were normal in SCD, but WM oxygen delivery was 35% lower than in controls. Age and hematocrit were the strongest predictors for WM CBF and oxygen delivery in patients with SCD. There was spatial co-localization between regions of low oxygen delivery and WM hyperintensities on T2 FLAIR imaging. To conclude, oxygen delivery is preserved in the GM of SCD patients, but is decreased throughout the WM, particularly in areas prone to WM silent strokes.

CEREBRAL BLOOD FLOW AND PREDICTORS OF WHITE MATTER LESIONS IN ADULTS WITH TETRALOGY OF FALLOT.

Long-term outcomes for Tetralogy of Fallot (TOF) have improved dramatically in recent years, but survivors are still afflicted by cerebral damage. In this paper, we characterized the prevalence and predictors of cerebral silent infarction (SCI) and their relationship to cerebral blood flow (CBF) in 46 adult TOF patients. We calculated both whole brain and regional CBF using 2D arterial spin labeling (ASL) images, and investigated the spatial overlap between voxel-wise CBF values and white matter hyperintensities (WMHs) identified from T2-FLAIR images. SCIs were found in 83% of subjects and were predicted by the year of the patient's first cardiac surgery and patient's age at scanning (combined r2 0.44). CBF was not different in brain regions prone to stroke compared with healthy white matter.

ORCHESTRAL FULLY CONVOLUTIONAL NETWORKS FOR SMALL LESION SEGMENTATION IN BRAIN MRI.

White matter (WM) lesion identification and segmentation has proved of clinical importance for diagnosis, treatment and neurological outcomes. Convolutional neural networks (CNN) have demonstrated their success for large lesion load segmentation, but are not sensitive to small deep WM and sub-cortical lesion segmentation. We propose to use multi-scale and supervised fully convolutional networks (FCN) to segment small WM lesions in 22 anemic patients. The multiple scales enable us to identify the small lesions while reducing many false alarms, and the multi-supervised scheme allows a better management of the unbalanced data. Compared to a single FCN (Dice score ~0.31), the performance on the testing dataset of our proposed networks achieved a Dice score of 0.78.

Pseudo continuous arterial spin labeling quantification in anemic subjects with hyperemic cerebral blood flow.

PURPOSE: To investigate possible sources of quantification errors in global cerebral blood flow (CBF) measurements by comparing pseudo continuous arterial spin labeling (PCASL) and phase contrast (PC) MRI in anemic, hyperemic subjects. METHODS: All studies were performed on a Philips 3T Achieva MRI scanner. PC and PCASL CBF examinations were performed in 10 healthy, young adult subjects and 18 young adults with chronic anemia syndromes including sickle cell disease and thalassemia. CBF estimates from single and two compartment ASL kinetic models were compared. Numerical simulation and flow phantom experiments were used to explore the effects of blood velocity and B1+ on CBF quantification and labeling efficiency. RESULTS: PCASL CBF underestimated PC in both populations using a single compartment model (30.1±9.2% control, 45.2±17.2% anemia). Agreement substantially improved using a two-compartment model (-8.0±6.0% control, 11.7±12.3% anemia). Four of the anemic subjects exhibited venous outflow of ASL signal, suggestive of cerebrovascular shunt, possibly confounding PC-PCASL comparisons. Additionally, sub-study experiments demonstrated that B1+ was diminished at the labeling plane (82.9±5.1%), resulting in suboptimal labeling efficiency. Correcting labeling efficiency for diminished B1+, PCASL slightly overestimated PC CBF in controls (-15.4±6.8%) and resulted in better matching of CBF estimates in anemic subjects (0.7±10.0% without outflow, 10.5±9.4% with outflow). CONCLUSIONS: This work demonstrates that a two-compartment model is critical for PCASL quantification in hyperemic subjects. Venous outflow and B1+ under-excitation may also contribute to flow underestimation, but further study of these effects is required.

Multivariate surface-based analysis of corpus callosum in patients with sickle cell disease.

Sickle cell disease (SCD) is a genetic hematological disease in which the hemoglobin molecule in red blood cells is abnormal. It is closely associated with many symptoms, including pain, anemia, chest syndrome and neurocognitive impairment. One of the most debilitating symptoms is elevated risk for cerebro-vascular accidents. The corpus callosum (CC), as the largest and most prominent white matter (WM) structure in the brain, can reflect the chronic cerebrovascular damage resulting from silent strokes or infarctions in asymptomatic SCD patients. While a lot of studies have reported WM alterations in this cohort, little is known about the shape deformation of the CC. Here we perform the first surface morphometry analysis of the CC in SCD patients using four different shape metrics on T1-weighted magnetic resonance images. We detect regional surface morphological differences in the CC between 11 patients and 10 healthy control subjects. Differences are located in the genu, posterior midbody and splenium, potentially casting light on the anatomical substrates underlying neuropsychological test differences between the SCD and control groups.

Tract specific analysis in patients with sickle cell disease.

Sickle cell disease (SCD) is a hereditary blood disorder in which the oxygen-carrying hemoglobin molecule in red blood cells is abnormal. It affects numerous people in the world and leads to a shorter life span, pain, anemia, serious infections and neurocognitive decline. Tract-Specific Analysis (TSA) is a statistical method to evaluate white matter alterations due to neurocognitive diseases, using diffusion tensor magnetic resonance images. Here, for the first time, TSA is used to compare 11 major brain white matter (WM) tracts between SCD patients and age-matched healthy subjects. Alterations are found in the corpus callosum (CC), the cortico-spinal tract (CST), inferior fronto-occipital fasciculus (IFO), inferior longitudinal fasciculus (ILF), superior longitudinal fasciculus (SLF), and uncinated fasciculus (UNC). Based on previous studies on the neurocognitive functions of these tracts, the significant areas found in this paper might be related to several cognitive impairments and depression, both of which are observed in SCD patients.