Mesiotemporal atrophy and hippocampal diffusivity distinguish amnestic from non-amnestic vascular cognitive impairment

A Deep Learning-Based Model for Classification of Different Subtypes of Subcortical Vascular Cognitive Impairment With FLAIR

Deep learning methods have shown their great capability of extracting high-level features from image and have been used for effective medical imaging classification recently. However, training samples of medical images are restricted by the amount of patients as well as medical ethics issues, making it hard to train the neural networks. In this paper, we propose a novel end-to-end three-dimensional (3D) attention-based residual neural network (ResNet) architecture to classify different subtypes of subcortical vascular cognitive impairment (SVCI) with single-shot T2-weighted fluid-attenuated inversion recovery (FLAIR) sequence. Our aim is to develop a convolutional neural network to provide a convenient and effective way to assist doctors in the diagnosis and early treatment of the different subtypes of SVCI. The experiment data in this paper are collected from 242 patients from the Neurology Department of Renji Hospital, including 78 amnestic mild cognitive impairment (a-MCI), 70 nonamnestic MCI (na-MCI), and 94 no cognitive impairment (NCI). The accuracy of our proposed model has reached 98.6% on a training set and 97.3% on a validation set. The test accuracy on an untrained testing set reaches 93.8% with robustness. Our proposed method can provide a convenient and effective way to assist doctors in the diagnosis and early treatment.

Large-scale network dysfunction in vascular cognitive disorder supports connectional diaschisis in advanced arteriosclerosis

BACKGROUND AND PURPOSE

The interrelation of cognitive performance, cerebrovascular damage and brain functional connectivity (FC) in advanced arteriosclerosis remains unclear. Our aim was to investigate the associations between FC, white matter damage and cognitive impairment in carotid artery disease.

METHODS

Seventy-one participants with a recent cerebrovascular event and with written informed consent underwent resting-state functional magnetic resonance imaging and the Addenbrooke's Cognitive Examination - Revised (ACE-R). Network and inter-hemispheric FC metrics were compared between cognitively normal and impaired subjects, and interrelated with cognition. In order to explore the nature of FC changes, their associations with microstructural damage of related white matter tracts and cognitive performance were investigated, followed by mediation analysis.

RESULTS

Participants with global cognitive impairment showed reduced FC compared to the cognitively intact subjects within the central executive network (CEN), and between hemispheres. Patients with executive dysfunction had decreased CEN FC whilst patients with memory loss demonstrated low FC in both the CEN and the default mode network (DMN). Global performance correlated with connectivity metrics of the CEN hub with DMN nodes, and between hemispheres. Cingulum mean diffusivity (MD) was negatively correlated with ACE-R and CEN-DMN FC. The cingulum MD-cognition association was partially mediated by CEN-DMN FC.

CONCLUSIONS

Long-range functional disconnection of the CEN with DMN nodes is the main feature of cognitive impairment in elderly subjects with symptomatic carotid artery disease. Our findings provide further support for the connectional diaschisis concept of vascular cognitive disorder, and highlight a mediation role of functional disconnection to explain associations between microstructural white matter tract damage and cognitive impairment.

Ongoing Secondary Degeneration of the Limbic System in Patients With Ischemic Stroke: A Longitudinal MRI Study

Purpose: Ongoing post-stroke structural degeneration and neuronal loss preceding neuropsychological symptoms such as cognitive decline and depression are poorly understood. Various substructures of the limbic system have been linked to cognitive impairment. In this longitudinal study, we investigated the post-stroke macro- and micro-structural integrity of the limbic system using structural and diffusion tensor magnetic resonance imaging.

Materials and Methods: Nineteen ischemic stroke patients (11 men, 8 women, average age 53.4 ± 12.3, range 18–75 years), with lesions remote from the limbic system, were serially imaged three times over 1 year. Structural and diffusion-tensor images (DTI) were obtained on a 3.0 T MRI system. The cortical thickness, subcortical volume, mean diffusivity (MD), and fractional anisotropy (FA) were measured in eight different regions of the limbic system. The National Institutes of Health Stroke Scale (NIHSS) was used for clinical assessment. A mixed model for multiple factors was used for statistical analysis, and p-values <0.05 was considered significant.

Results: All patients demonstrated improved NIHSS values over time. The ipsilesional subcortical volumes of the thalamus, hippocampus, and amygdala significantly decreased (p < 0.05) and MD significantly increased (p < 0.05). The ipsilesional cortical thickness of the entorhinal and perirhinal cortices was significantly smaller than the contralesional hemisphere at 12 months (p < 0.05). The cortical thickness of the cingulate gyrus at 12 months was significantly decreased at the caudal and isthmus regions as compared to the 1 month assessment (p < 0.05). The cingulum fibers had elevated MD at the ipsilesional caudal-anterior and posterior regions compared to the corresponding contralesional regions.

Conclusion: Despite the decreasing NIHSS scores, we found ongoing unilateral neuronal loss/secondary degeneration in the limbic system, irrespective of the lesion location. These results suggest a possible anatomical basis for post stroke psychiatric complications.

Progress toward standardized diagnosis of vascular cognitive impairment: Guidelines from the Vascular Impairment of Cognition Classification Consensus Study

INTRODUCTION

Progress in understanding and management of vascular cognitive impairment (VCI) has been hampered by lack of consensus on diagnosis, reflecting the use of multiple different assessment protocols. A large multinational group of clinicians and researchers participated in a two-phase Vascular Impairment of Cognition Classification Consensus Study (VICCCS) to agree on principles (VICCCS-1) and protocols (VICCCS-2) for diagnosis of VCI. We present VICCCS-2.

METHODS

We used VICCCS-1 principles and published diagnostic guidelines as points of reference for an online Delphi survey aimed at achieving consensus on clinical diagnosis of VCI.

RESULTS

Six survey rounds comprising 65-79 participants agreed guidelines for diagnosis of VICCCS-revised mild and major forms of VCI and endorsed the National Institute of Neurological Disorders-Canadian Stroke Network neuropsychological assessment protocols and recommendations for imaging.

DISCUSSION

The VICCCS-2 suggests standardized use of the National Institute of Neurological Disorders-Canadian Stroke Network recommendations on neuropsychological and imaging assessment for diagnosis of VCI so as to promote research collaboration.