Deep artificial neural networks (DNNs) have moved to the forefront of medical image analysis due to their success in classification, segmentation, and detection challenges. A principal challenge in ...large-scale deployment of DNNs in neuroimage analysis is the potential for shifts in signal-to-noise ratio, contrast, resolution, and presence of artifacts from site to site due to variances in scanners and acquisition protocols. DNNs are famously susceptible to these distribution shifts in computer vision. Currently, there are no benchmarking platforms or frameworks to assess the robustness of new and existing models to specific distribution shifts in MRI, and accessible multi-site benchmarking datasets are still scarce or task-specific. To address these limitations, we propose ROOD-MRI: a novel platform for benchmarking the Robustness of DNNs to Out-Of-Distribution (OOD) data, corruptions, and artifacts in MRI. This flexible platform provides modules for generating benchmarking datasets using transforms that model distribution shifts in MRI, implementations of newly derived benchmarking metrics for image segmentation, and examples for using the methodology with new models and tasks. We apply our methodology to hippocampus, ventricle, and white matter hyperintensity segmentation in several large studies, providing the hippocampus dataset as a publicly available benchmark. By evaluating modern DNNs on these datasets, we demonstrate that they are highly susceptible to distribution shifts and corruptions in MRI. We show that while data augmentation strategies can substantially improve robustness to OOD data for anatomical segmentation tasks, modern DNNs using augmentation still lack robustness in more challenging lesion-based segmentation tasks. We finally benchmark U-Nets and vision transformers, finding robustness susceptibility to particular classes of transforms across architectures. The presented open-source platform enables generating new benchmarking datasets and comparing across models to study model design that results in improved robustness to OOD data and corruptions in MRI.
Display omitted
•Developed open-source benchmarking platform and metrics for robustness of DNNs.•Quantified sensitivity of DNNs to OOD data on three neuroimaging segmentation tasks.•Modern CNNs are highly susceptible to distribution shift, corruptions and artifacts.•Simple augmentation strategies improve robustness for anatomical segmentation tasks.•Vision transformers exhibit improved robustness over FCNs.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Plasma amyloid beta (Aβ) and tau are emerging as accessible biomarkers for Alzheimer's disease (AD). However, many assays exist with variable test performances, highlighting the need for a ...comparative assessment to identify the most valid assays for future use in AD and to apply to other settings in which the same biomarkers may be useful, namely, cerebral amyloid angiopathy (CAA). CAA is a progressive cerebrovascular disease characterized by deposition of Aβ40 and Aβ42 in cortical and leptomeningeal vessels. Novel immunotherapies for AD can induce amyloid‐related imaging abnormalities resembling CAA‐related inflammation. Few studies have evaluated plasma biomarkers in CAA. Identifying a CAA signature could facilitate diagnosis, prognosis, and a safer selection of patients with AD for emerging immunotherapies. This review evaluates studies that compare the diagnostic test performance of plasma biomarker techniques in AD and cerebrovascular and plasma biomarker profiles of CAA; it also discusses novel hypotheses and future avenues for plasma biomarker research in CAA.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
Studies suggest a role of the innate immune system, including the activity of neutrophils, in neurodegeneration related to Alzheimer's disease (AD), but prospective cognitive data remain lacking in ...humans. We aimed to investigate the predictive relationship between neutrophil-associated inflammatory proteins in peripheral blood and changes in memory and executive function over 1 year in patients with AD.
Participants with AD were identified from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Neutrophil gelatinase-associated lipocalin (NGAL), myeloperoxidase (MPO), interleukin-8 (IL-8), macrophage inflammatory protein-1 beta (MIP-1β), and tumor necrosis factor (TNF) were assayed by luminex immunofluorescence multiplex assay at baseline. Confirmatory factor analysis was used to test an underlying neutrophil associated plasma inflammatory factor. Composite z-scores for memory and executive function were generated from multiple tests at baseline and at 1 year. A multiple linear regression model was used to investigate the association of the baseline inflammatory factor with changes in memory and executive function over 1 year.
Among AD patients (n = 109, age = 74.8 ± 8.1, 42% women, Mini Mental State Examination MMSE = 23.6 ± 1.9), the neutrophil-related inflammatory proteins NGAL (λ = 0.595, p < .001), MPO (λ = 0.575, p < .001), IL-8 (λ = 0.525, p < .001), MIP-1β (λ = 0.411, p = .008), and TNF (λ = 0.475, p < .001) were found to inform an underlying factor. Over 1 year, this inflammatory factor predicted a decline in executive function (β = - 0.152, p = 0.015) but not memory (β = 0.030, p = 0.577) in models controlling for demographics, brain atrophy, white matter hyperintensities, the ApoE ε4 allele, concomitant medications, and baseline cognitive performance.
An inflammatory factor constructed from five neutrophil-related markers in peripheral blood predicted a decline in executive function over 1 year in people with mild AD.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
This article investigates whether increasing mandatory educational attainment through compulsory schooling legislation encourages women to delay childbearing. We use variation induced by changes in ...compulsory schooling laws in both the US and Norway to estimate the effect in two very different institutional environments. We find evidence that increased compulsory schooling does in fact reduce the incidence of teenage childbearing in both the US and Norway, and these estimates are quite robust to various specification checks. These results suggest that legislation aimed at improving educational outcomes may have spillover effects onto the fertility decisions of teenagers.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, INZLJ, IZUM, KILJ, NLZOH, NMLJ, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBMB, UL, UM, UPUK, ZRSKP
This scientific statement provides an overview of the evidence on vascular contributions to cognitive impairment and dementia. Vascular contributions to cognitive impairment and dementia of later ...life are common. Definitions of vascular cognitive impairment (VCI), neuropathology, basic science and pathophysiological aspects, role of neuroimaging and vascular and other associated risk factors, and potential opportunities for prevention and treatment are reviewed. This statement serves as an overall guide for practitioners to gain a better understanding of VCI and dementia, prevention, and treatment.
Writing group members were nominated by the writing group co-chairs on the basis of their previous work in relevant topic areas and were approved by the American Heart Association Stroke Council Scientific Statement Oversight Committee, the Council on Epidemiology and Prevention, and the Manuscript Oversight Committee. The writing group used systematic literature reviews (primarily covering publications from 1990 to May 1, 2010), previously published guidelines, personal files, and expert opinion to summarize existing evidence, indicate gaps in current knowledge, and, when appropriate, formulate recommendations using standard American Heart Association criteria. All members of the writing group had the opportunity to comment on the recommendations and approved the final version of this document. After peer review by the American Heart Association, as well as review by the Stroke Council leadership, Council on Epidemiology and Prevention Council, and Scientific Statements Oversight Committee, the statement was approved by the American Heart Association Science Advisory and Coordinating Committee.
The construct of VCI has been introduced to capture the entire spectrum of cognitive disorders associated with all forms of cerebral vascular brain injury-not solely stroke-ranging from mild cognitive impairment through fully developed dementia. Dysfunction of the neurovascular unit and mechanisms regulating cerebral blood flow are likely to be important components of the pathophysiological processes underlying VCI. Cerebral amyloid angiopathy is emerging as an important marker of risk for Alzheimer disease, microinfarction, microhemorrhage and macrohemorrhage of the brain, and VCI. The neuropathology of cognitive impairment in later life is often a mixture of Alzheimer disease and microvascular brain damage, which may overlap and synergize to heighten the risk of cognitive impairment. In this regard, magnetic resonance imaging and other neuroimaging techniques play an important role in the definition and detection of VCI and provide evidence that subcortical forms of VCI with white matter hyperintensities and small deep infarcts are common. In many cases, risk markers for VCI are the same as traditional risk factors for stroke. These risks may include but are not limited to atrial fibrillation, hypertension, diabetes mellitus, and hypercholesterolemia. Furthermore, these same vascular risk factors may be risk markers for Alzheimer disease. Carotid intimal-medial thickness and arterial stiffness are emerging as markers of arterial aging and may serve as risk markers for VCI. Currently, no specific treatments for VCI have been approved by the US Food and Drug Administration. However, detection and control of the traditional risk factors for stroke and cardiovascular disease may be effective in the prevention of VCI, even in older people.
Vascular contributions to cognitive impairment and dementia are important. Understanding of VCI has evolved substantially in recent years, based on preclinical, neuropathologic, neuroimaging, physiological, and epidemiological studies. Transdisciplinary, translational, and transactional approaches are recommended to further our understanding of this entity and to better characterize its neuropsychological profile. There is a need for prospective, quantitative, clinical-pathological-neuroimaging studies to improve knowledge of the pathological basis of neuroimaging change and the complex interplay between vascular and Alzheimer disease pathologies in the evolution of clinical VCI and Alzheimer disease. Long-term vascular risk marker interventional studies beginning as early as midlife may be required to prevent or postpone the onset of VCI and Alzheimer disease. Studies of intensive reduction of vascular risk factors in high-risk groups are another important avenue of research.
Neuroplasticity accompanying learning is a key mediator of stroke rehabilitation. Training in playing music in healthy populations and patients with movement disorders requires resources within ...motor, sensory, cognitive, and affective systems, and coordination among these systems. We investigated effects of music‐supported therapy (MST) in chronic stroke on motor, cognitive, and psychosocial functions compared to conventional physical training (GRASP). Twenty‐eight adults with unilateral arm and hand impairment were randomly assigned to MST (n = 14) and GRASP (n = 14) and received 30 h of training over a 10‐week period. The assessment was conducted at four time points: before intervention, after 5 weeks, after 10 weeks, and 3 months after training completion. As for two of our three primary outcome measures concerning motor function, all patients slightly improved in Chedoke–McMaster Stroke Assessment hand score, while the time to complete Action Research Arm Test became shorter in the MST group. The third primary outcome measure for well‐being, Stroke Impact Scale, was improved for emotion and social communication earlier in MST and coincided with the improved executive function for task switching and music rhythm perception. The results confirmed previous findings and expanded the potential usage of MST for enhancing quality of life in community‐dwelling chronic‐stage survivors.
Neuroplasticity accompanying learning is a key mediator of stroke rehabilitation. Training in playing music in healthy populations and patients with movement disorders, requires resources within motor, sensory, cognitive, and affective systems, and coordination among these systems. We investigated effects of music‐supported therapy (MST) in chronic stroke on motor, cognitive, and psychosocial functions compared to conventional physical training (GRASP).
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, IZUM, KILJ, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBMB, UL, UM, UPUK
Although the brain lacks conventional lymphatic vessels found in peripheral tissue, evidence suggests that the space surrounding the vasculature serves a similar role in the clearance of fluid and ...metabolic waste from the brain. With aging, neurodegeneration, and cerebrovascular disease, these microscopic perivascular spaces can become enlarged, allowing for visualization and quantification on structural MRI. The purpose of this review is to: (i) describe some of the recent pre-clinical findings from basic science that shed light on the potential neurophysiological mechanisms driving glymphatic and perivascular waste clearance, (ii) review some of the pathobiological etiologies that may lead to MRI-visible enlarged perivascular spaces (ePVS), (iii) describe the possible clinical implications of ePVS, (iv) evaluate existing qualitative and quantitative techniques used for measuring ePVS burden, and (v) propose future avenues of research that may improve our understanding of this potential clinical neuroimaging biomarker for fluid and metabolic waste clearance dysfunction in neurodegenerative and neurovascular diseases.
Full text
Available for:
EMUNI, FZAB, GEOZS, IJS, IMTLJ, KILJ, KISLJ, MFDPS, NUK, OILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UL, UM, UPUK, VKSCE, ZAGLJ
•We examined the efficacy and safety of nabilone as a treatment for agitation in patients with moderate-to-severe Alzheimer's disease.•Nabilone was shown to improve agitation, overall behavior, and ...caregiver distress compared to placebo. While sedation was greater in nabilone treatment group, there were no between-group differences in treatment-limiting sedation.•A larger/longer trial is needed to confirm the safety and efficacy of nabilone for agitation.
To investigate the efficacy and safety of nabilone for agitation in patients with moderate-to-severe Alzheimer's disease (AD).
This 14-week randomized double-blind crossover trial compared nabilone to placebo (6 weeks each) with a 1-week washout between phases.
Patients were recruited from a long-term care facility and geriatric psychiatry clinics.
Patients had AD (standardized Mini-Mental State Examination sMMSE ≤24) and agitation (Neuropsychiatric Inventory-Nursing Home version NPI-NH-agitation/aggression subscore ≥3).
Nabilone (target 1–2 mg) versus placebo.
The primary outcome was agitation (Cohen Mansfield Agitation Inventory CMAI). Secondary outcomes included NPI-NH total, NPI-NH caregiver distress, cognition (sMMSE and Severe Impairment Battery SIB or Alzheimer's Disease Assessment Scale of Cognition), global impression (Clinician's Global Impression of Change CGIC), and adverse events.
Thirty-nine patients (mean ± SD age = 87 ± 10, sMMSE = 6.5 ± 6.8, CMAI = 67.9 ± 17.6, NPI-NH total = 34.3 ± 15.8, 77% male, nabilone dose = 1.6 ± 0.5 mg) were randomized. There were no crossover or treatment-order effects. Using a linear mixed model, treatment differences (95% CI) in CMAI (b = −4.0 −6.5 to −1.5, t(30.2) = −3.3, p = 0.003), NPI-NH total (b = −4.6 −7.5 to −1.6, t(32.9) = −3.1, p = 0.004), NPI-NH caregiver distress (b = −1.7 −3.4 to −0.07, t(33.7) = −2.1, p = 0.041), and sMMSE (b = 1.1 0.1–2.0, t(22.6) = 2.4, p = 0.026) all favored nabilone. However, in those who completed the SIB (n = 25) treatment differences favored placebo (b = −4.6 −7.3 to −1.8, t(20.7) = −4.8, p = 0.003). CGIC improvement during nabilone (47%) and placebo (23%) was not significantly different (McNemar's test, exact p = 0.09). There was more sedation during nabilone (45%) compared to placebo (16%) phases (McNemar's test, exact p = 0.02), but treatment-limiting sedation was not significantly different (McNemar's test, exact p = 0.22).
Nabilone may be an effective treatment for agitation. However, sedation and cognition should be closely monitored.
BACKGROUND:Several sets of diagnostic criteria have been published for vascular dementia since the 1960s. The continuing ambiguity in vascular dementia definition warrants a critical reexamination.
...METHODS:Participants at a special symposium of the International Society for Vascular Behavioral and Cognitive Disorders (VASCOG) in 2009 critiqued the current criteria. They drafted a proposal for a new set of criteria, later reviewed through multiple drafts by the group, including additional experts and the members of the Neurocognitive Disorders Work Group of the fifth revision of Diagnostic and Statistical Manual (DSM-5) Task Force.
RESULTS:Cognitive disorders of vascular etiology are a heterogeneous group of disorders with diverse pathologies and clinical manifestations, discussed broadly under the rubric of vascular cognitive disorders (VCD). The continuum of vascular cognitive impairment is recognized by the categories of Mild Vascular Cognitive Disorder, and Vascular Dementia or Major Vascular Cognitive Disorder. Diagnostic thresholds are defined. Clinical and neuroimaging criteria are proposed for establishing vascular etiology. Subtypes of VCD are described, and the frequent cooccurrence of Alzheimer disease pathology emphasized.
CONCLUSIONS:The proposed criteria for VCD provide a coherent approach to the diagnosis of this diverse group of disorders, with a view to stimulating clinical and pathologic validation studies. These criteria can be harmonized with the DSM-5 criteria such that an international consensus on the criteria for VCD may be achieved.
Studies evaluating individuals for endothelial injury after endovascular thrombectomy (EVT) have been done by means of retrieved human thrombus, MR vessel-wall imaging, and animal histopathological ...studies. These techniques have limitations, because MR imaging has insufficient spatial resolution to directly visualize endothelium, and histopathological examinations are performed ex vivo and are unable to provide real-time patterns of injury. The purpose of the current study was to obtain in vivo intraluminal imaging after EVT by using optical coherence tomography (OCT), examining for evidence of endothelial injury in real time.Three consecutive patients with acute basilar artery occlusion underwent OCT imaging immediately after EVT. There were no complications and adequate images were obtained for all patients. Anatomical features of the vessel wall were discernible, including intima, media, adventitia, and internal/external elastic lamina. Basilar artery thick concentric plaque fibrosis was present, causing outward remodeling and loss of the internal/external lamina in certain regions. Evidence of significant residual thrombus was also visible, with mostly red thrombus present despite complete angiographic revascularization. The residual thrombus was not visible on CT, MR, or cerebral angiography and could certainly cause ongoing function-limiting strokes with occlusion of adjacent vital basilar perforators after EVT.
PDF
