Generative Model of Brain Microbleeds for MRI Detection of Vascular Marker of Neurodegenerative Diseases.
Ontology highlight
ABSTRACT:
SUBMITTER: Momeni S
PROVIDER: S-EPMC8716785 | biostudies-literature | 2021
REPOSITORIES: biostudies-literature
Publications
Generative Model of Brain Microbleeds for MRI Detection of Vascular Marker of Neurodegenerative Diseases.
Momeni Saba S Fazlollahi Amir A Lebrat Leo L Yates Paul P Rowe Christopher C Gao Yongsheng Y Liew Alan Wee-Chung AW Salvado Olivier O
Frontiers in neuroscience 20211216
Cerebral microbleeds (CMB) are increasingly present with aging and can reveal vascular pathologies associated with neurodegeneration. Deep learning-based classifiers can detect and quantify CMB from MRI, such as susceptibility imaging, but are challenging to train because of the limited availability of ground truth and many confounding imaging features, such as vessels or infarcts. In this study, we present a novel generative adversarial network (GAN) that has been trained to generate three-dime ...[more]
Similar Datasets
Project description:The treatment of neurodegenerative diseases is hindered by lack of interventions capable of steering multimodal whole-brain dynamics towards patterns indicative of preserved brain health. To address this problem, we combined deep learning with a model capable of reproducing whole-brain functional connectivity in patients diagnosed with Alzheimer's disease (AD) and behavioral variant frontotemporal dementia (bvFTD). These models included disease-specific atrophy maps as priors to modulate local parameters, revealing increased stability of hippocampal and insular dynamics as signatures of brain atrophy in AD and bvFTD, respectively. Using variational autoencoders, we visualized different pathologies and their severity as the evolution of trajectories in a low-dimensional latent space. Finally, we perturbed the model to reveal key AD- and bvFTD-specific regions to induce transitions from pathological to healthy brain states. Overall, we obtained novel insights on disease progression and control by means of external stimulation, while identifying dynamical mechanisms that underlie functional alterations in neurodegeneration.
| S-EPMC10063230 | biostudies-literature
Project description:The incidence of neurodegenerative diseases in the older population has increased in recent years. A considerable number of studies have been performed to characterize these diseases. Imaging analysis is an important biomarker for the diagnosis of neurodegenerative disease. Objective and reliable assessment and precise detection are important for the early diagnosis of neurodegenerative diseases. Artificial intelligence (AI) using brain MRI applied to the study of neurodegenerative diseases could promote early diagnosis and optimal decisions for treatment plans. MRI-based AI software have been developed and studied worldwide. Representatively, there are MRI-based volumetry and segmentation software. In this review, we present the development process of brain volumetry analysis software in neurodegenerative diseases, currently used and developed AI software for neurodegenerative disease in the Republic of Korea, probable uses of AI in the future, and AI software limitations.
| S-EPMC9514516 | biostudies-literature
Project description:Cerebral microbleeds, observed as small, spherical hypointense regions on gradient echo (GRE) or susceptibility weighted (SWI) magnetic resonance imaging (MRI) sequences, reflect small hemorrhagic infarcts, and are associated with conditions such as vascular dementia, small vessel disease, cerebral amyloid angiopathy, and Alzheimer's disease. The current gold standard for detecting and rating cerebral microbleeds in a research context is visual inspection by trained raters, a process that is both time consuming and subject to poor reliability. We present here a novel method to automate microbleed detection on GRE and SWI images. We demonstrate in a community-based cohort of older adults that the method is highly sensitive (greater than 92% of all microbleeds accurately detected) across both modalities, with reasonable precision (fewer than 20 and 10 false positives per scan on GRE and SWI, respectively). We also demonstrate that the algorithm can be used to identify microbleeds over longitudinal scans with a higher level of sensitivity than visual ratings (50% of longitudinal microbleeds correctly labeled by the algorithm, while manual ratings was 30% or lower). Further, the algorithm identifies the anatomical localization of microbleeds based on brain atlases, and greatly reduces time spent completing visual ratings (43% reduction in visual rating time). Our automatic microbleed detection instrument is ideal for implementation in large-scale studies that include cross-sectional and longitudinal scanning, as well as being capable of performing well across multiple commonly used MRI modalities.
| S-EPMC7889861 | biostudies-literature
Project description:Traumatic brain injury (TBI) is a leading cause of death and disability in people under 45. Advanced imaging techniques to identify injury and classify severity in the first few hours and days following trauma could improve patient stratification and aid clinical decision making. Traumatic cerebral microbleeds (TCMBs), detectable on magnetic resonance susceptibility weighted imaging (SWI), can be used as markers of long-term clinical outcome. However, the relationship between TCMBs and injury severity in the first few hours after injury, and their natural evolution, is unknown.We obtained SWI scans in 10 healthy controls, and 13 patients scanned 3-24h following TBI and again at 7-15days. TCMBs were identified and total volume quantified for every lesion in each scan.TCMBs were present in 6 patients, all with more severe injury classified by GCS. No lesions were identified in patients with an initial GCS of 15. Improvement in GCS in the first 15days following injury was significantly associated with a reduction in microbleed volume over the same time-period.MRI is feasible in severely injured patients in the first 24h after trauma. Detection of TCMBs using SWI provides an objective early marker of injury severity following trauma. TCMBs revealed in this time frame, offer the potential to help determine the degree of injury, improving stratification, in order to identify patients who require admission to hospital, transfer to a specialist center, or an extended period of intubation on intensive care.
| S-EPMC5541760 | biostudies-other
Project description:Neurodevelopmental and neurodegenerative diseases (NDDs) with severe neurological/psychiatric symptoms, such as cerebrovascular pathology in AD, CAA, and chronic stroke, have brought greater attention with their incidence and prevalence having markedly increased over the past few years. Causes of the significant neuropathologies, especially those observed in neurological diseases in the CNS, are commonly believed to involve multiple factors such as an age, a total environment, genetics, and an immunity contributing to their progression, neuronal, and vascular injuries. We primarily focused on the studies of glial involvement/dysfunction in part with the blood-brain barrier (BBB) and the neurovascular unit (NVU) changes, and the vascular mechanisms, which have been both suggested as critical roles in chronic stroke and many other NDDs. It has been noted that glial cells including astrocytes (which outnumber other cell types in the CNS) essentially contribute more to the BBB integrity, extracellular homeostasis, neurotransmitter release, regulation of neurogenic niches in response to neuroinflammatory stimulus, and synaptic plasticity. In a recent study for NDDs utilizing cellular and molecular biology and genetic and pharmacological tools, the role of reactive astrocytes (RACs) and gliosis was demonstrated, able to trigger pathophysiological/psychopathological detrimental changes during the disease progression. We speculate, in particular, the BBB, the NVU, and changes of the astrocytes (potentially different populations from the RACs) not only interfere with neuronal development and synaptogenesis, but also generate oxidative damages, contribute to beta-amyloid clearances and disrupted vasculature, as well as lead to neuroinflammatory disorders. During the past several decades, stem cell therapy has been investigated with a research focus to target related neuro-/vascular pathologies (cell replacement and repair) and neurological/psychiatric symptoms (paracrine protection and homeostasis). Evidence shows that transplantation of neurogenic or vasculogenic cells could be achieved to pursue differentiation and maturation within the diseased brains as expected. It would be hoped that, via regulating functions of astrocytes, astrocytic involvement, and modulation of the BBB, the NVU and astrocytes should be among major targets for therapeutics against NDDs pathogenesis by drug and cell-based therapies. The non-invasive strategies in combination with stem cell transplantation such as the well-tested intranasal deliveries for drug and stem cells by our and many other groups show great translational potentials in NDDs. Neuroimaging and clinically relevant analyzing tools need to be evaluated in various NDDs brains.
| S-EPMC8032950 | biostudies-literature
Project description:BACKGROUND:Minimally invasive specific biomarkers of neurodegenerative diseases (NDs) would facilitate patient selection and disease progression monitoring. We describe the assessment of circulating brain-enriched microRNAs as potential biomarkers for Alzheimer's disease (AD), frontotemporal dementia (FTD), Parkinson's disease (PD), and amyotrophic lateral sclerosis (ALS). METHODS:In this case-control study, the plasma samples were collected from 250 research participants with a clinical diagnosis of AD, FTD, PD, and ALS, as well as from age- and sex-matched control subjects (n = 50 for each group), recruited from 2003 to 2015 at the University of Pennsylvania Health System, including the Alzheimer's Disease Center, the Parkinson's Disease and Movement Disorders Center, the Frontotemporal Degeneration Center, and the Amyotrophic Lateral Sclerosis Clinic. Each group was randomly divided into training and confirmation sets of equal size. To evaluate the potential of circulating microRNAs enriched in specific brain regions affected by NDs and present in synapses as biomarkers of NDs, the levels of 37 brain-enriched and inflammation-associated microRNAs in the plasma of all participants were measured using individual qRT-PCR. A "microRNA pair" approach was used for data normalization. RESULTS:MicroRNA pairs and their combinations (classifiers) capable of differentiating NDs from control and from each other were defined using independently and jointly analyzed training and confirmation datasets. AD, PD, FTD, and ALS are differentiated from control with accuracy of 0.89, 0.90, 0.88, and 0.83 (AUCs, 0.96, 0.96, 0.94, and 0.93), respectively; NDs are differentiated from each other with accuracy ranging from 0.77 (AUC, 0.87) for AD vs. FTD to 0.93 (AUC, 0.98) for AD vs. ALS. The data further indicate sex dependence of some microRNA markers. The average increase in accuracy in distinguishing ND from control for all and male/female groups is 0.06; the largest increase is for ALS, from 0.83 for all participants to 0.92/0.98 for male/female participants. CONCLUSIONS:The work presented here suggests the possibility of developing microRNA-based diagnostics for detection and differentiation of NDs. Larger multicenter clinical studies are needed to further evaluate circulating brain-enriched microRNAs as biomarkers for NDs and to investigate their association with other ND biomarkers in clinical trial settings.
| S-EPMC5679501 | biostudies-literature
Project description:Traumatic microbleeds are small foci of hypointensity seen on T2*-weighted MRI in patients following head trauma that have previously been considered a marker of axonal injury. The linear appearance and location of some traumatic microbleeds suggests a vascular origin. The aims of this study were to: (i) identify and characterize traumatic microbleeds in patients with acute traumatic brain injury; (ii) determine whether appearance of traumatic microbleeds predict clinical outcome; and (iii) describe the pathology underlying traumatic microbleeds in an index patient. Patients presenting to the emergency department following acute head trauma who received a head CT were enrolled within 48 h of injury and received a research MRI. Disability was defined using Glasgow Outcome Scale-Extended ≤6 at follow-up. All magnetic resonance images were interpreted prospectively and were used for subsequent analysis of traumatic microbleeds. Lesions on T2* MRI were stratified based on 'linear' streak-like or 'punctate' petechial-appearing traumatic microbleeds. The brain of an enrolled subject imaged acutely was procured following death for evaluation of traumatic microbleeds using MRI targeted pathology methods. Of the 439 patients enrolled over 78 months, 31% (134/439) had evidence of punctate and/or linear traumatic microbleeds on MRI. Severity of injury, mechanism of injury, and CT findings were associated with traumatic microbleeds on MRI. The presence of traumatic microbleeds was an independent predictor of disability (P < 0.05; odds ratio = 2.5). No differences were found between patients with punctate versus linear appearing microbleeds. Post-mortem imaging and histology revealed traumatic microbleed co-localization with iron-laden macrophages, predominately seen in perivascular space. Evidence of axonal injury was not observed in co-localized histopathological sections. Traumatic microbleeds were prevalent in the population studied and predictive of worse outcome. The source of traumatic microbleed signal on MRI appeared to be iron-laden macrophages in the perivascular space tracking a network of injured vessels. While axonal injury in association with traumatic microbleeds cannot be excluded, recognizing traumatic microbleeds as a form of traumatic vascular injury may aid in identifying patients who could benefit from new therapies targeting the injured vasculature and secondary injury to parenchyma.
| S-EPMC6821371 | biostudies-literature
Project description:Triage is essential for the early diagnosis and reporting of neurologic emergencies. Herein, we report the development of an anomaly detection algorithm (ADA) with a deep generative model trained on brain computed tomography (CT) images of healthy individuals that reprioritizes radiology worklists and provides lesion attention maps for brain CT images with critical findings. In the internal and external validation datasets, the ADA achieved area under the curve values (95% confidence interval) of 0.85 (0.81-0.89) and 0.87 (0.85-0.89), respectively, for detecting emergency cases. In a clinical simulation test of an emergency cohort, the median wait time was significantly shorter post-ADA triage than pre-ADA triage by 294 s (422.5 s [interquartile range, IQR 299] to 70.5 s [IQR 168]), and the median radiology report turnaround time was significantly faster post-ADA triage than pre-ADA triage by 297.5 s (445.0 s [IQR 298] to 88.5 s [IQR 179]) (all p < 0.001).
| S-EPMC9307758 | biostudies-literature
Project description:IntroductionLittle is known about the epidemiology of brain microbleeds in racially/ethnically diverse populations.MethodsIn the Multi-Ethnic Study of Atherosclerosis, brain microbleeds were identified from 3T magnetic resonance imaging susceptibility-weighted imaging sequences using deep learning models followed by radiologist review.ResultsAmong 1016 participants without prior stroke (25% Black, 15% Chinese, 19% Hispanic, 41% White, mean age 72), microbleed prevalence was 20% at age 60 to 64.9 and 45% at ≥85 years. Deep microbleeds were associated with older age, hypertension, higher body mass index, and atrial fibrillation, and lobar microbleeds with male sex and atrial fibrillation. Overall, microbleeds were associated with greater white matter hyperintensity volume and lower total white matter fractional anisotropy.DiscussionResults suggest differing associations for lobar versus deep locations. Sensitive microbleed quantification will facilitate future longitudinal studies of their potential role as an early indicator of vascular pathology.
| S-EPMC11163989 | biostudies-literature
Project description:Finite Element (FE) models of brain mechanics have improved our understanding of the brain response to rapid mechanical loads that produce traumatic brain injuries. However, these models have rarely incorporated vasculature, which limits their ability to predict the response of vessels to head impacts. To address this shortcoming, here we used high-resolution MRI scans to map the venous system anatomy at a submillimetre resolution. We then used this map to develop an FE model of veins and incorporated it in an anatomically detailed FE model of the brain. The model prediction of brain displacement at different locations was compared to controlled experiments on post-mortem human subject heads, yielding over 3,100 displacement curve comparisons, which showed fair to excellent correlation between them. We then used the model to predict the distribution of axial strains and strain rates in the veins of a rugby player who had small blood deposits in his white matter, known as microbleeds, after sustaining a head collision. We hypothesised that the distribution of axial strain and strain rate in veins can predict the pattern of microbleeds. We reconstructed the head collision using video footage and multi-body dynamics modelling and used the predicted head accelerations to load the FE model of vascular injury. The model predicted large axial strains in veins where microbleeds were detected. A region of interest analysis using white matter tracts showed that the tract group with microbleeds had 95th percentile peak axial strain and strain rate of 0.197 and 64.9 s-1 respectively, which were significantly larger than those of the group of tracts without microbleeds (0.163 and 57.0 s-1). This study does not derive a threshold for the onset of microbleeds as it investigated a single case, but it provides evidence for a link between strain and strain rate applied to veins during head impacts and structural damage and allows for future work to generate threshold values. Moreover, our results suggest that the FE model has the potential to be used to predict intracranial vascular injuries after TBI, providing a more objective tool for TBI assessment and improving protection against it.
| S-EPMC9065595 | biostudies-literature