Prediction of cognitive decline in Parkinson's disease based on MRI radiomics and clinical features: A multicenter study.

OBJECTIVE: To develop and validate a multimodal combinatorial model based on whole-brain magnetic resonance imaging (MRI) radiomic features for predicting cognitive decline in patients with Parkinson's disease (PD). METHODS: This study included a total of 222 PD patients with normal baseline cognition, of whom 68 had cognitive impairment during a 4-year follow-up period. All patients underwent MRI scans, and radiomic features were extracted from the whole-brain MRI images of the training set, and dimensionality reduction was performed to construct a radiomics model. Subsequently, Screening predictive factors for cognitive decline from clinical features and then combining those with a radiomics model to construct a multimodal combinatorial model for predicting cognitive decline in PD patients. Evaluate the performance of the comprehensive model using the receiver-operating characteristic curve, confusion matrix, F1 score, and survival curve. In addition, the quantitative characteristics of diffusion tensor imaging (DTI) from corpus callosum were selected from 52 PD patients to further validate the clinical efficacy of the model. RESULTS: The multimodal combinatorial model has good classification performance, with areas under the curve of 0.842, 0.829, and 0.860 in the training, test, and validation sets, respectively. Significant differences were observed in the number of cognitive decline PD patients and corpus callosum-related DTI parameters between the low-risk and high-risk groups distinguished by the model (p < 0.05). The survival curve analysis showed a statistically significant difference in the progression time of mild cognitive impairment between the low-risk and the high-risk groups. CONCLUSIONS: The building of a multimodal combinatorial model based on radiomic features from MRI can predict cognitive decline in PD patients, thus providing adaptive strategies for clinical practice.

Functional neuroimaging as a catalyst for integrated neuroscience.

Functional magnetic resonance imaging (fMRI) enables non-invasive access to the awake, behaving human brain. By tracking whole-brain signals across a diverse range of cognitive and behavioural states or mapping differences associated with specific traits or clinical conditions, fMRI has advanced our understanding of brain function and its links to both normal and atypical behaviour. Despite this headway, progress in human cognitive neuroscience that uses fMRI has been relatively isolated from rapid advances in other subdomains of neuroscience, which themselves are also somewhat siloed from one another. In this Perspective, we argue that fMRI is well-placed to integrate the diverse subfields of systems, cognitive, computational and clinical neuroscience. We first summarize the strengths and weaknesses of fMRI as an imaging tool, then highlight examples of studies that have successfully used fMRI in each subdomain of neuroscience. We then provide a roadmap for the future advances that will be needed to realize this integrative vision. In this way, we hope to demonstrate how fMRI can help usher in a new era of interdisciplinary coherence in neuroscience.

Imaging Markers of Vascular Brain Health: Quantification, Clinical Implications, and Future Directions.

Cerebrovascular disease (CVD) manifests through a broad spectrum of mechanisms that negatively impact brain and cognitive health. Oftentimes, CVD changes (excluding acute stroke) are insufficiently considered in aging and dementia studies which can lead to an incomplete picture of the etiologies contributing to the burden of cognitive impairment. Our goal with this focused review is 3-fold. First, we provide a research update on the current magnetic resonance imaging methods that can measure CVD lesions as well as early CVD-related brain injury specifically related to small vessel disease. Second, we discuss the clinical implications and relevance of these CVD imaging markers for cognitive decline, incident dementia, and disease progression in Alzheimer disease, and Alzheimer-related dementias. Finally, we present our perspective on the outlook and challenges that remain in the field. With the increased research interest in this area, we believe that reliable CVD imaging biomarkers for aging and dementia studies are on the horizon.

Dissection of brain-wide resting-state and functional somatosensory circuits by fMRI with optogenetic silencing.

To further advance functional MRI (fMRI)-based brain science, it is critical to dissect fMRI activity at the circuit level. To achieve this goal, we combined brain-wide fMRI with neuronal silencing in well-defined regions. Since focal inactivation suppresses excitatory output to downstream pathways, intact input and suppressed output circuits can be separated. Highly specific cerebral blood volume-weighted fMRI was performed with optogenetic stimulation of local GABAergic neurons in mouse somatosensory regions. Brain-wide spontaneous somatosensory networks were found mostly in ipsilateral cortical and subcortical areas, which differed from the bilateral homotopic connections commonly observed in resting-state fMRI data. The evoked fMRI responses to somatosensory stimulation in regions of the somatosensory network were successfully dissected, allowing the relative contributions of spinothalamic (ST), thalamocortical (TC), corticothalamic (CT), corticocortical (CC) inputs, and local intracortical circuits to be determined. The ventral posterior thalamic nucleus receives ST inputs, while the posterior medial thalamic nucleus receives CT inputs from the primary somatosensory cortex (S1) with TC inputs. The secondary somatosensory cortex (S2) receives mostly direct CC inputs from S1 and a few TC inputs from the ventral posterolateral nucleus. The TC and CC input layers in cortical regions were identified by laminar-specific fMRI responses with a full width at half maximum of <150 µm. Long-range synaptic inputs in cortical areas were amplified approximately twofold by local intracortical circuits, which is consistent with electrophysiological recordings. Overall, whole-brain fMRI with optogenetic inactivation revealed brain-wide, population-based, long-range circuits, which could complement data typically collected in conventional microscopic functional circuit studies.

Fusion of Preoperative and Postoperative Imaging May Predict the Diagnostic Yield of Stereotactic Needle Brain Biopsies.

BACKGROUND: Stereotactic needle brain biopsy is a commonly used neurosurgical procedure. However, up to 15% of biopsies result in undiagnostic pathology reports. Repeat biopsy or continued management without a diagnosis are often considered after undiagnostic biopsies. There have been no reports about the role of postoperative imaging in predicting the diagnostic yield of stereotactic biopsies. METHODS: We retrospectively assessed all stereotactic biopsies performed over an 11-year period. We performed fusion of immediate postoperative computed tomography (CT) with preoperative MRI, to document whether the air bubble in the postoperative CT was located within the targeted lesion. We then evaluated the association of this fusion-based accuracy assessment with the diagnostic yield of the biopsy. RESULTS: Fewer than 5% of biopsies did not have an air bubble on postoperative CT. A total of 226 biopsies were performed for 219 patients. In our sample, 213 of 226 biopsies were accurate (94.2% accuracy rate), and 203 of 226 biopsies gave a definitive diagnosis (89.8% diagnostic rate). In those cases where the fusion was accurate, the diagnostic rate was 93.9%. When the fusion was inaccurate, the diagnostic rate was only 23.1% (odds ratio 51.5, 95% confidence interval 12.6-210.44, P < 0.001). Of all patient, imaging, surgical, and admission parameters, the only parameter that correlated with diagnostic outcome of the biopsy was the fusion construct accuracy. CONCLUSIONS: Fusion of immediate postoperative CT with preoperative imaging is predictive of the diagnostic rate. In cases where the pathology report following a biopsy is not diagnostic, this fusion may be useful in making decisions regarding repeat biopsy or considering other diagnostic options.

Longitudinal Changes of Sensorimotor Resting-State Functional Connectivity Differentiate between Patients with Thalamic Infarction and Pontine Infarction.

Purpose. We investigated the disparate influence of lesion location on functional damage and reorganization of the sensorimotor brain network in patients with thalamic infarction and pontine infarction. Methods. Fourteen patients with unilateral infarction of the thalamus and 14 patients with unilateral infarction of the pons underwent longitudinal fMRI measurements and motor functional assessment five times during a 6-month period (<7 days, at 2 weeks, 1 month, 3 months, and 6 months after stroke onset). Twenty-five age- and sex-matched controls underwent MRI examination across five consecutive time points in 6 months. Functional images from patients with left hemisphere lesions were first flipped from the left to the right side. The voxel-wise connectivity analyses between the reference time course of each ROI (the contralateral dorsal lateral putamen (dl-putamen), pons, ventral anterior (VA), and ventral lateral (VL) nuclei of the thalamus) and the time course of each voxel in the sensorimotor area were performed for all five measurements. One-way ANOVA was used to identify between-group differences in functional connectivity (FC) at baseline stage (<7 days after stroke onset), with infarction volume included as a nuisance variable. The family-wise error (FWE) method was used to account for multiple comparison issues using SPM software. Post hoc repeated-measure ANOVA was applied to examine longitudinal FC reorganization. Results. At baseline stage, significant differences were detected between the contralateral VA and ipsilateral postcentral gyrus (cl_VA-ip_postcentral), contralateral VL and ipsilateral precentral gyrus (cl_VL-ip_precentral). Repeated measures ANOVA revealed that the FC change of cl_VA-ip_postcentral differ significantly among the three groups over time. The significant changes of FC between cl_VA and ip_postcentral at different time points in the thalamic infarction group showed that compared with 7 days after stroke onset, there was significantly increased FC of cl_VA-ip_postcentral at 1 month, 3 months, and 6 months after stroke onset. Conclusions. The different patterns of sensorimotor functional damage and reorganization in patients with pontine infarction and thalamic infarction may provide insights into the neural mechanisms underlying functional recovery after stroke.

Clinical applications of magnetic resonance imaging based functional and structural connectivity.

Advances in computational neuroimaging techniques have expanded the armamentarium of imaging tools available for clinical applications in clinical neuroscience. Non-invasive, in vivo brain MRI structural and functional network mapping has been used to identify therapeutic targets, define eloquent brain regions to preserve, and gain insight into pathological processes and treatments as well as prognostic biomarkers. These tools have the real potential to inform patient-specific treatment strategies. Nevertheless, a realistic appraisal of clinical utility is needed that balances the growing excitement and interest in the field with important limitations associated with these techniques. Quality of the raw data, minutiae of the processing methodology, and the statistical models applied can all impact on the results and their interpretation. A lack of standardization in data acquisition and processing has also resulted in issues with reproducibility. This limitation has had a direct impact on the reliability of these tools and ultimately, confidence in their clinical use. Advances in MRI technology and computational power as well as automation and standardization of processing methods, including machine learning approaches, may help address some of these issues and make these tools more reliable in clinical use. In this review, we will highlight the current clinical uses of MRI connectomics in the diagnosis and treatment of neurological disorders; balancing emerging applications and technologies with limitations of connectivity analytic approaches to present an encompassing and appropriate perspective.

Altered substrate metabolism in neurodegenerative disease: new insights from metabolic imaging.

Neurodegenerative diseases (NDs), such as Alzheimer's disease (AD), Parkinson's disease (PD) and multiple sclerosis (MS), are relatively common and devastating neurological disorders. For example, there are 6 million individuals living with AD in the United States, a number that is projected to grow to 14 million by the year 2030. Importantly, AD, PD and MS are all characterized by the lack of a true disease-modifying therapy that is able to reverse or halt disease progression. In addition, the existing standard of care for most NDs only addresses the symptoms of the disease. Therefore, alternative strategies that target mechanisms underlying the neuropathogenesis of disease are much needed. Recent studies have indicated that metabolic alterations in neurons and glia are commonly observed in AD, PD and MS and lead to changes in cell function that can either precede or protect against disease onset and progression. Specifically, single-cell RNAseq studies have shown that AD progression is tightly linked to the metabolic phenotype of microglia, the key immune effector cells of the brain. However, these analyses involve removing cells from their native environment and performing measurements in vitro, influencing metabolic status. Therefore, technical approaches that can accurately assess cell-specific metabolism in situ have the potential to be transformative to our understanding of the mechanisms driving AD. Here, we review our current understanding of metabolism in both neurons and glia during homeostasis and disease. We also evaluate recent advances in metabolic imaging, and discuss how emerging modalities, such as fluorescence lifetime imaging microscopy (FLIM) have the potential to determine how metabolic perturbations may drive the progression of NDs. Finally, we propose that the temporal, regional, and cell-specific characterization of brain metabolism afforded by FLIM will be a critical first step in the rational design of metabolism-focused interventions that delay or even prevent NDs.

Current and Emerging Tools for Hepatocellular Carcinoma Surveillance.

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide. Early detection of HCC enables patients to avail curative therapies that can improve patient survival. Current international guidelines advocate for the enrollment of patients at high risk for HCC, like those with cirrhosis, in surveillance programs that perform ultrasound every 6 months. In recent years, many studies have further characterized the utility of established screening strategies and have introduced new promising tools for HCC surveillance. In this review, we provide an overview of the most promising new imaging modalities and biomarkers for the detection of HCC. We discuss the role of imaging tools like ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI) in the early detection of HCC, and describe recent innovations which can potentially enhance their applicability, including contrast enhanced ultrasound, low-dose CT scans, and abbreviated MRI. Next, we outline the data supporting the use of three circulating biomarkers (i.e., alpha-fetoprotein [AFP], AFP lens culinaris agglutinin-reactive fraction, and des-gamma-carboxy prothrombin) in HCC surveillance, and expand on multiple emerging liquid biopsy biomarkers, including methylated cell-free DNA (cfDNA), cfDNA mutations, extracellular vesicles, and circulating tumor cells. These promising new imaging modalities and biomarkers have the potential to improve early detection, and thus improve survival, in patients with HCC.

Predictability of polygenic risk score for progression to dementia and its interaction with APOE ε4 in mild cognitive impairment.

BACKGROUND: The combinatorial effect of multiple genetic factors calculated as a polygenic risk score (PRS) has been studied to predict disease progression to Alzheimer's disease (AD) from mild cognitive impairment (MCI). Previous studies have investigated the performance of PRS in the prediction of disease progression to AD by including and excluding single nucleotide polymorphisms within the region surrounding the APOE gene. These studies may have missed the APOE genotype-specific predictability of PRS for disease progression to AD. METHODS: We analyzed 732 MCI from the Alzheimer's Disease Neuroimaging Initiative cohort, including those who progressed to AD within 5 years post-baseline (n = 270) and remained stable as MCI (n = 462). The predictability of PRS including and excluding the APOE region (PRS+APOE and PRS-APOE) on the conversion to AD and its interaction with the APOE ε4 carrier status were assessed using Cox regression analyses. RESULTS: PRS+APOE (hazard ratio [HR] 1.468, 95% CI 1.335-1.615) and PRS-APOE (HR 1.293, 95% CI 1.157-1.445) were both associated with a significantly increased risk of MCI progression to dementia. The interaction between PRS+APOE and APOE ε4 carrier status was significant with a P-value of 0.0378. The association of PRSs with the progression risk was stronger in APOE ε4 non-carriers (PRS+APOE: HR 1.710, 95% CI 1.244-2.351; PRS-APOE: HR 1.429, 95% CI 1.182-1.728) than in APOE ε4 carriers (PRS+APOE: HR 1.167, 95% CI 1.005-1.355; PRS-APOE: HR 1.172, 95% CI 1.020-1.346). CONCLUSIONS: PRS could predict the conversion of MCI to dementia with a stronger association in APOE ε4 non-carriers than APOE ε4 carriers. This indicates PRS as a potential genetic predictor particularly for MCI with no APOE ε4 alleles.

Smart magnetic resonance imaging-based theranostics for cancer.

Smart theranostics are dynamic platforms that integrate multiple functions, including at least imaging, therapy, and responsiveness, in a single agent. This review showcases a variety of responsive theranostic agents developed specifically for magnetic resonance imaging (MRI), due to the privileged position this non-invasive, non-ionising imaging modality continues to hold within the clinical imaging field. Different MRI smart theranostic designs have been devised in the search for more efficient cancer therapy, and improved diagnostic efficiency, through the increase of the local concentration of therapeutic effectors and MRI signal intensity in pathological tissues. This review explores novel small-molecule and nanosized MRI theranostic agents for cancer that exhibit responsiveness to endogenous (change in pH, redox environment, or enzymes) or exogenous (temperature, ultrasound, or light) stimuli. The challenges and obstacles in the design and in vivo application of responsive theranostics are also discussed to guide future research in this interdisciplinary field towards more controllable, efficient, and diagnostically relevant smart theranostics agents.

PET Imaging for Gynecologic Malignancies.

This review article summarizes the clinical applications of established and emerging PET tracers in the evaluation of the 5 most common gynecologic malignancies: endometrial, ovarian, cervical, vaginal, and vulvar cancers. Emphasis is given to 2-deoxy-2-[18F]fluoro-d-glucose as the most widely used and studied tracer, with additional clinical tracers also explored. The common imaging protocols are discussed, including standard dose ranges and uptake times, established roles, as well as the challenges and future directions of these imaging techniques. The key points are emphasized with images from selected cases.

Clinical Applications of PET/MR Imaging.

PET/MR imaging is in routine clinical use and is at least as effective as PET/CT for oncologic and neurologic studies with advantages with certain PET radiopharmaceuticals and applications. In addition, whole body PET/MR imaging substantially reduces radiation dosages compared with PET/CT which is particularly relevant to pediatric and young adult population. For cancer imaging, assessment of hepatic, pelvic, and soft-tissue malignancies may benefit from PET/MR imaging. For neurologic imaging, volumetric brain MR imaging can detect regional volume loss relevant to cognitive impairment and epilepsy. In addition, the single-bed position acquisition enables dynamic brain PET imaging without extending the total study length which has the potential to enhance the diagnostic information from PET.

Neuroimaging in Frontotemporal Dementia: Heterogeneity and Relationships with Underlying Neuropathology.

Frontotemporal dementia encompasses a group of clinical syndromes defined pathologically by degeneration of the frontal and temporal lobes. Historically, these syndromes have been challenging to diagnose, with an average of about three years between the time of symptom onset and the initial evaluation and diagnosis. Research in the field of neuroimaging has revealed numerous biomarkers of the various frontotemporal dementia syndromes, which has provided clinicians with a method of narrowing the differential diagnosis and improving diagnostic accuracy. As such, neuroimaging is considered a core investigative tool in the evaluation of neurodegenerative disorders. Furthermore, patterns of neurodegeneration correlate with the underlying neuropathological substrates of the frontotemporal dementia syndromes, which can aid clinicians in determining the underlying etiology and improve prognostication. This review explores the advancements in neuroimaging and discusses the phenotypic and pathologic features of behavioral variant frontotemporal dementia, semantic variant primary progressive aphasia, and nonfluent variant primary progressive aphasia, as seen on structural magnetic resonance imaging and positron emission tomography.

Fully Automatic Adaptive Meshing Based Segmentation of the Ventricular System for Augmented Reality Visualization and Navigation.

OBJECTIVE: Effective image segmentation of cerebral structures is fundamental to 3-dimensional techniques such as augmented reality. To be clinically viable, segmentation algorithms should be fully automatic and easily integrated in existing digital infrastructure. We created a fully automatic adaptive-meshing-based segmentation system for T1-weighted magnetic resonance images (MRI) to automatically segment the complete ventricular system, running in a cloud-based environment that can be accessed on an augmented reality device. This study aims to assess the accuracy and segmentation time of the system by comparing it to a manually segmented ground truth dataset. METHODS: A ground truth (GT) dataset of 46 contrast-enhanced and non-contrast-enhanced T1-weighted MRI scans was manually segmented. These scans also were uploaded to our system to create a machine-segmented (MS) dataset. The GT data were compared with the MS data using the Sørensen-Dice similarity coefficient and 95% Hausdorff distance to determine segmentation accuracy. Furthermore, segmentation times for all GT and MS segmentations were measured. RESULTS: Automatic segmentation was successful for 45 (98%) of 46 cases. Mean Sørensen-Dice similarity coefficient score was 0.83 (standard deviation [SD] = 0.08) and mean 95% Hausdorff distance was 19.06 mm (SD = 11.20). Segmentation time was significantly longer for the GT group (mean = 14405 seconds, SD = 7089) when compared with the MS group (mean = 1275 seconds, SD = 714) with a mean difference of 13,130 seconds (95% confidence interval 10,130-16,130). CONCLUSIONS: The described adaptive meshing-based segmentation algorithm provides accurate and time-efficient automatic segmentation of the ventricular system from T1 MRI scans and direct visualization of the rendered surface models in augmented reality.

Perfusion Imaging and Clinical Outcome in Acute Ischemic Stroke with Large Core.

OBJECTIVE: Mechanical thrombectomy (MT) is not recommended for acute stroke with large vessel occlusion (LVO) and a large volume of irreversibly injured tissue ("core"). Perfusion imaging may identify a subset of patients with large core who benefit from MT. METHODS: We compared two cohorts of LVO-related patients with large core (>50 ml on diffusion-weighted-imaging or CT-perfusion using RAPID), available perfusion imaging, and treated within 6 hours from onset by either MT + Best Medical Management (BMM) in one prospective study, or BMM alone in the pre-MT era from a prospective registry. Primary outcome was 90-day modified Rankin Scale ≤2. We searched for an interaction between treatment group and amount of penumbra as estimated by the mismatch ratio (MMRatio = critical hypoperfusion/core volume). RESULTS: Overall, 107 patients were included (56 MT + BMM and 51 BMM): Mean age was 68 ± 15 years, median core volume 99 ml (IQR: 72-131) and MMRatio 1.4 (IQR: 1.0-1.9). Baseline clinical and radiological variables were similar between the two groups, except for a higher intravenous thrombolysis rate in the BMM group. The MMRatio strongly modified the clinical outcome following MT (pinteraction < 0.001 for continuous MMRatio); MT was associated with a higher rate of good outcome in patients with, but not in those without, MMRatio>1.2 (adjusted OR [95% CI] = 6.8 [1.7-27.0] vs 0.7 [0.1-6.2], respectively). Similar findings were present for MMRatio ≥1.8 in the subgroup with core ≥70 ml. Parenchymal hemorrhage on follow-up imaging was more frequent in the MT + BMM group regardless of the MMRatio. INTERPRETATION: Perfusion imaging may help select which patients with large core should be considered for MT. Randomized studies are warranted. ANN NEUROL 2021;90:417-427.

Spinal Fluid Myeloid Microvesicles Predict Disease Course in Multiple Sclerosis.

OBJECTIVE: In vivo measures of myeloid activity are promising biomarkers in multiple sclerosis. We previously demonstrated that cerebrospinal fluid (CSF) myeloid microvesicles are markers of microglial/macrophage activity and neuroinflammation in multiple sclerosis. Here, we aimed at investigating the diagnostic and prognostic value of myeloid microvesicles in a clinical setting. METHODS: Six hundred one patients discharged with a diagnosis of neuroinflammatory, neurodegenerative, or no neurological disease were enrolled. Myeloid microvesicles were measured with flow cytometry as isolectin B4-positive events in fresh CSF. Clinical, demographical, and magnetic resonance imaging (MRI) data were collected at diagnosis (all patients) and during follow-up (n = 176). RESULTS: CSF myeloid microvesicles were elevated in neuroinflammatory patients compared to the neurodegenerative and control groups. In multiple sclerosis, microvesicles were higher in patients with MRI disease activity and their concentration increased along with the number of enhancing lesions (p < 0.0001, Jonckheere-Terpstra test). CSF myeloid microvesicles were also higher in patients with higher disease activity in the month and year preceding diagnosis. Microvesicles excellently discriminated between the relapsing-remitting and control groups (receiver operator characteristic curve, area under the curve = 0.939, p < 0.0001) and between radiologically isolated syndrome and unspecific brain lesions (0.942, p < 0.0001). Furthermore, microvesicles were independent predictors of prognosis for both the relapsing-remitting and progressive groups. Microvesicles independently predicted future disease activity in relapsing-remitting patients (hazard ratio [HR] = 1.967, 95% confidence interval [CI] = 1.147-3.372), correcting for prognostic factors of standard clinical use. In the progressive group, microvesicles were independent predictors of disability accrual (HR = 10.767, 95% CI = 1.335-86.812). INTERPRETATION: Our results confirm that CSF myeloid microvesicles are a clinically meaningful biomarker of neuroinflammation and microglial/macrophage activity in vivo. These findings may support a possible use in clinical practice during diagnostic workup and prognostic assessment. ANN NEUROL 2021;90:253-265.

Longitudinal Changes in Hippocampal Network Connectivity in Alzheimer's Disease.

OBJECTIVE: The hippocampus is connected to 2 distinct cortical brain networks, the posterior-medial and the anterior-temporal networks, involving different medial temporal lobe (MTL) subregions. The aim of this study was to assess the functional alterations of these 2 networks, their changes over time, and links to cognition in Alzheimer's disease. METHODS: We assessed MTL connectivity in 53 amyloid-ß-positive patients with mild cognitive impairment and AD dementia and 68 healthy elderly controls, using resting-state functional magnetic resonance imaging, cross-sectionally and longitudinally. First, we compared the functional connectivity of the posterior-medial and anterior-temporal networks within the control group to highlight their specificities. Second, we compared the connectivity of these networks between groups, and between baseline and 18-month follow-up in patients. Third, we assessed the association in the connectivity changes between the 2 networks, and with cognitive performance. RESULTS: We found decreased connectivity in patients specifically between the hippocampus and the posterior-medial network, together with increased connectivity between several MTL subregions and the anterior-temporal network. Moreover, changes in the posterior-medial and anterior-temporal networks were interrelated such that decreased MTL-posterior-medial connectivity was associated with increased MTL-anterior-temporal connectivity. Finally, both MTL-posterior-medial decrease and MTL-anterior-temporal increase predicted cognitive decline. INTERPRETATION: Our findings demonstrate that longitudinal connectivity changes in the posterior-medial and anterior-temporal hippocampal networks are linked together and that they both contribute to cognitive decline in Alzheimer's disease. These results shed light on the critical role of the posterior-medial and anterior-temporal networks in Alzheimer's disease pathophysiology and clinical symptoms. ANN NEUROL 2021;90:391-406.

Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction.

Quantitative susceptibility mapping (QSM) and R2* mapping are MRI post-processing methods that quantify tissue magnetic susceptibility and transverse relaxation rate distributions. However, QSM and R2* acquisitions are relatively slow, even with parallel imaging. Incoherent undersampling and compressed sensing reconstruction techniques have been used to accelerate traditional magnitude-based MRI acquisitions; however, most do not recover the full phase signal, as required by QSM, due to its non-convex nature. In this study, a learning-based Deep Complex Residual Network (DCRNet) is proposed to recover both the magnitude and phase images from incoherently undersampled data, enabling high acceleration of QSM and R2* acquisition. Magnitude, phase, R2*, and QSM results from DCRNet were compared with two iterative and one deep learning methods on retrospectively undersampled acquisitions from six healthy volunteers, one intracranial hemorrhage and one multiple sclerosis patients, as well as one prospectively undersampled healthy subject using a 7T scanner. Peak signal to noise ratio (PSNR), structural similarity (SSIM), root-mean-squared error (RMSE), and region-of-interest susceptibility and R2* measurements are reported for numerical comparisons. The proposed DCRNet method substantially reduced artifacts and blurring compared to the other methods and resulted in the highest PSNR, SSIM, and RMSE on the magnitude, R2*, local field, and susceptibility maps. Compared to two iterative and one deep learning methods, the DCRNet method demonstrated a 3.2% to 9.1% accuracy improvement in deep grey matter susceptibility when accelerated by a factor of four. The DCRNet also dramatically shortened the reconstruction time of single 2D brain images from 36-140 seconds using conventional approaches to only 15-70 milliseconds.

Proportions of circulating transitional B cells associate with MRI activity in interferon beta-treated multiple sclerosis patients.

B-cells contribute to MS pathogenesis. The association of circulating B-cell phenotypes with combined unique active lesions (CUA) on MRI at 48 weeks follow-up was investigated in 50 interferon beta-treated MS patients. Transitional B-cell proportions were lower in participants with CUA at week 0 and 48 [p = 0.004, p = 0.002]. A decrease in circulating anti-EBNA-1 IgG levels between week 0 and 48 associated with absence of CUA [p = 0.047], but not with B-cell profiles. In a multi-factor model for CUA-risk, transitional B-cell proportions contributed independent from NK/T-cell ratio, change in anti-EBNA-1 IgG, and vitamin D supplementation. Transitional B-cells may predict treatment response in MS.