Search

Early detection of unruptured intracranial aneurysms (UIAs) enables better rupture risk and preventative treatment assessment. UIAs are usually diagnosed on Time-of-Flight Magnetic Resonance Angiographs (TOF-MRA) or contrast-enhanced Computed Tomography Angiographs (CTA). Various automatic voxel-based deep learning UIA detection methods have been developed, but these are limited to a single modality. We propose a modality-independent UIA detection method using a geometric deep learning model with high resolution surface meshes of brain vessels. A mesh convolutional neural network with ResU-Net style architecture was used. UIA detection performance was investigated with different input and pooling mesh resolutions, and including additional edge input features (shape index and curvedness). Both a higher resolution mesh (15,000 edges) and additional curvature edge features improved performance (average sensitivity: 65.6%, false positive count/image (FPC/image): 1.61). UIAs were detected in an independent TOF-MRA test set and a CTA test set with average sensitivity of 52.0% and 48.3% and average FPC/image of 1.04 and 1.05 respectively. We provide modality-independent UIA detection using a deep-learning vascular surface mesh model with comparable performance to state-of-the-art UIA detection methods.

•Is there an association between white matter hyperintensities lesion location and language?•In this cross-sectional analysis of a population-based study that included 819 persons, the left cingulum of the cingulate gyrus was identified as a strategic white matter tract for the Modified Boston Naming Test.•The language performance is sensitive to ischemic damage in a larger subcortical network than was previously thought.

IMPORTANCE: Neuropathology studies show a high prevalence of cortical microinfarcts (CMIs) in aging individuals, especially in patients with cerebrovascular disease and dementia. However, most, are invisible on T1- and T2-weighted magnetic resonance imaging (MRI), raising the question of how to explain this mismatch. Studies on small acute infarcts, detected on diffusion-weighted imaging (DWI), suggest that infarcts are largest in their acute phase and reduce in size thereafter. Therefore, we hypothesized that a subset of the CMI that are invisible on MRI can be detected on MRI in their acute phase. However, to our knowledge, a serial imaging study investigating the temporal dynamics of acute CMI (A-CMI) is lacking. OBJECTIVE: To determine the prevalence of chronic CMI (C-CMI) and the cumulative incidence and temporal dynamics of A-CMI in individuals with cerebral small vessel disease (SVD). DESIGN, SETTING, PARTICIPANTS AND EXPOSURES: The RUN DMC—Intense study is a single-center hospital-based prospective cohort study on SVD performed between March 2016 and November 2017 and comprising 10 monthly 3-T MRI scans, including high-resolution DWI, 3-dimensional T1, 3-dimensional fluid-attenuated inversion recovery, and T2. One hundred six individuals from the previous longitudinal RUN DMC study were recruited based on the presence of progression of white matter hyperintensities on MRI between 2006 and 2015 and exclusion of causes of cerebral ischemia other than SVD. Fifty-four individuals (50.9%) participated. The median total follow-up duration was 39.5 weeks (interquartile range, 37.8-40.3). Statistical data analysis was performed between May and October 2019. MAIN OUTCOMES AND MEASURES: We determined the prevalence of C-CMI using the baseline T1, fluid-attenuated inversion recovery, and T2 scans. Monthly high-resolution DWI scans (n = 472) were screened to determine the cumulative incidence of A-CMI. The temporal dynamics of A-CMI were determined based on the MRI scans collected during the first follow-up visit after A-CMI onset and the last available follow-up visit. RESULTS: The median age of the cohort at baseline MRI was 69 years (interquartile range, 66-74 years) and 34 participants (63%) were men. The prevalence of C-CMI was 35% (95% CI, 0.24-0.49). Monthly DWI detected 21 A-CMI in 7 of 54 participants, resulting in a cumulative incidence of 13% (95% CI, 0.06-0.24). All A-CMI disappeared on follow-up MRI. CONCLUSIONS AND RELEVANCE: Acute CMI never evolved into chronically MRI-detectable lesions. We suggest that these A-CMI underlie part of the submillimeter C-CMI encountered on neuropathological examination and thereby provide a source for the high CMI burden on neuropathology.

Lesion-symptom mapping (LSM) is used to capture the impact of lesions on cognitive performance while accounting for their location in the brain and is highly relevant for vascular cognitive impairment [1], [2]. Current LSM methods only consider a single lesion type and single cognitive score at a time. Neural networks (NNs) allow for multiple inputs (lesions) and outputs (cognitive domains), which might take interrelations of vascular lesions and cognitive subscores into account. Explainable AI (XAI) can be used to compute attribution maps reflecting which image locations are deemed important for a NN, even at an individual level. We explore the feasibility of NNs and XAI for LSM by comparing two NNs with current gold standard, support vector regression (SVR) [3], [4].

Deficits in the ability to distinguish between the left and right side of the body can severely impair daily life functioning. The current study examined the relation between left right orientation (LRO) impairments and somatosensory related deficits, ranging from primary somatosensory impairments to body representation impairments, in patients who suffered a recent stroke. We also examined which areas in the brain are associated with LRO impairments using a Voxel-based Lesion Symptom Mapping (VLSM) analysis.

The objectives of this study were to explore the possibilities of 7 T brain magnetic resonance imaging to visualize perivascular spaces (PVS) and to depict their related blood vessels.

Supplemental Digital Content is available in the text.

Cerebral small vessel diseases (SVDs) are a major cause of stroke and dementia. Yet, specific treatment strategies are lacking in part because of a limited understanding of the underlying disease processes. There is therefore an urgent need to study SVDs at their core, the small vessels themselves.

Disturbances of semantic and phonemic fluency are common after brain damage, as a manifestation of language, executive, or memory dysfunction. Lesion-symptom mapping (LSM) studies can provide fundamental insights in shared and distinct anatomical correlates of these cognitive functions and help to understand which patients suffer from these deficits. We performed a multivariate support vector regression-based lesion-symptom mapping and structural disconnection study on semantic and phonemic fluency in 1231 patients with acute ischemic stroke. With the largest-ever LSM study on verbal fluency we achieved almost complete brain lesion coverage. Lower performance on both fluency types was related to left hemispheric frontotemporal and parietal cortical regions, and subcortical regions centering on the left thalamus. Distinct correlates for phonemic fluency were the anterior divisions of middle and inferior frontal gyri. Distinct correlates for semantic fluency were the posterior regions of the middle and inferior temporal gyri, parahippocampal and fusiform gyri and triangular part of the inferior frontal gyrus. The disconnectome-based analyses additionally revealed phonemic fluency was associated with a more extensive frontoparietal white matter network, whereas semantic fluency was associated with disconnection of the fornix, mesiotemporal white matter, splenium of the corpus callosum. These results provide the most detailed outline of the anatomical correlates of phonemic and semantic fluency to date, stress the crucial role of subcortical regions and reveal a novel dissociation in the left temporal lobe.

Depression is the most common neuropsychiatric complication after stroke. Infarct location is associated with poststroke depressive symptoms (PSDS), but it remains debated which brain structures are critically involved. We performed a large-scale lesion-symptom mapping study to identify infarct locations and white matter disconnections associated with PSDS.