Background Among asymptomatic patients with severe carotid artery stenosis but no recent stroke or transient cerebral ischaemia, either carotid artery stenting (CAS) or carotid endarterectomy (CEA) can restore patency and reduce long-term stroke risks. However, from recent national registry data, each option causes about 1% procedural risk of disabling stroke or death. Comparison of their long-term protective effects requires large-scale randomised evidence.Methods ACST-2 is an international multicentre randomised trial of CAS versus CEA among asymptomatic patients with severe stenosis thought to require intervention, interpreted with all other relevant trials. Patients were eligible if they had severe unilateral or bilateral carotid artery stenosis and both doctor and patient agreed that a carotid procedure should be undertaken, but they were substantially uncertain which one to choose. Patients were randomly allocated to CAS or CEA and followed up at 1 month and then annually, for a mean 5 years. Procedural events were those within 30 days of the intervention. Intention-to-treat analyses are provided. Analyses including procedural hazards use tabular methods. Analyses and meta-analyses of non-procedural strokes use Kaplan-Meier and log-rank methods. The trial is registered with the ISRCTN registry, ISRCTN21144362.

The reliability of frontal and temporal fMRI activations for the determination of hemisphere language dominance was evaluated in comparison with intracarotid amytal testing (IAT). Twenty-two patients were studied by IAT (bilateral in 13, unilateral in 9 patients) and fMRI using a paradigm requiring semantic decisions. Global and regional (frontal and temporoparietal) lateralisation indices (LI) were calculated from the number of activated (r>0.4) voxels in both hemispheres. Frontolateral activations associated with the language task were seen in all patients, temporoparietal activations in 20 of 22. Regional LI corresponded better with IAT results than global LI. Frontolateral LI were consistent with IAT in all patients with bilateral IAT (including three patients with right dominant and one patient with bilateral language representation) and were not conflicting in any of the patients with unilateral IAT. Temporoparietal LI were discordant with IAT in two patients with atypical language representation. In the determination of hemisphere dominance for language, regional analysis of fMRI activation is superior to global analysis. In cases with clear-cut fMRI lateralisation, i.e. consistent lateralised activation of frontal and temporoparietal language zones, IAT may be unnecessary. FMRI should be performed prior to IAT in all patients going to be operated in brain regions potentially involved in language.

When searching for information, a human reader first glances over a document, spots relevant sections and then focuses on a few sentences for resolving her intention. However, the high variance of document structure complicates to identify the salient topic of a given section at a glance. To tackle this challenge, we present SECTOR, a model to support machine reading systems by segmenting documents into coherent sections and assigning topic labels to each section. Our deep neural network architecture learns a latent topic embedding over the course of a document. This can be leveraged to classify local topics from plain text and segment a document at topic shifts. In addition, we contribute WikiSection, a publicly available dataset with 242k labeled sections in English and German from two distinct domains: diseases and cities. From our extensive evaluation of 20 architectures, we report a highest score of 71.6% F1 for the segmentation and classification of 30 topics from the English city domain, scored by our SECTOR LSTM model with bloom filter embeddings and bidirectional segmentation. This is a significant improvement of 29.5 points F1 compared to state-of-the-art CNN classifiers with baseline segmentation. 1 Our source code is available under the Apache License 2.0 at https

Visual dysfunction is commonplace in schizophrenia and occurs alongside cognitive, psychotic and affective symptoms of the disorder. Psychophysical evidence suggests that this dysfunction results from impairments in the integration of low-level neural signals into complex cortical representations, which may also be associated with symptom formation. Despite the symptoms of schizophrenia occurring in a range of disorders, the integration deficit has not been tested in broader patient populations. Moreover, it remains unclear whether such deficits generalize across other sensory modalities. The present study assessed patients with a range of psychotic and nonpsychotic disorders and healthy controls on visual contrast detection, visual motion integration, auditory tone detection and auditory tone integration. The sample comprised a total of 249 participants (schizophrenia spectrum disorder n=98; bipolar affective disorder n=35; major depression n=31; other psychiatric conditions n=31; and healthy controls n=54), of whom 178 completed one or more visual task and 71 completed auditory tasks. Compared with healthy controls and nonpsychotic patients, psychotic patients trans-diagnostically were impaired on both visual and auditory integration, but unimpaired in simple visual or auditory detection. Impairment in visual motion integration was correlated with the severity of positive symptoms, and could not be accounted for by a reduction in processing speed, inattention or medication effects. Our results demonstrate that impaired sensory integration is not specific to schizophrenia, as has previously been assumed. Instead, sensory deficits are closely related to the presence of positive symptoms independent of diagnosis. The finding that equivalent integrative sensory processing is impaired in audition is consistent with hypotheses that propose a generalized deficit of neural integration in psychotic disorders.

Our objective was to evaluate tumor hemodynamics of paragangliomas of the head and neck. A 2D magnetic resonance projection angiography (MRPA) protocol was developed. Six patients with histologically proven paragangliomas were included in the study. The sequence protocol and data post-processing procedure were optimized in view of a high temporal resolution and maximum contrast-to-noise ratio. Image-time series of two freely orientated slabs could be obtained. Correlation analysis was performed to generate selective arterial and venous weighted angiographic images. Glomus tumors showed a rapid and intense homogeneous enhancement following the intravenous administration of contrast material. The smallest tumors investigated measured 10 mm and could be visualized clearly. Time-resolved analysis of the tumor enhancement revealed an early bolus arrival within the vascular bed of the tumor and prolonged incomplete washout indicating blood pooling. The dynamic contrast enhancement of glomus tumors can be evaluated non-invasively with MRPA providing a high temporal resolution and high image quality. The characteristic contrast enhancement of glomus tumors can be helpful in the diagnostic workup of lesions that may mimic glomus tumors.