Background and Purpose—
Studies on the prevalence and risk factors of white matter lesions (WMLs) in Tibetans living at high altitudes are scarce. We conducted this study to determine the prevalence and risks of WMLs in Tibetan patients without or with nonacute stroke.
Methods—
We undertook a retrospective analysis of medical records of patients treated at the People’s Hospital of Tibetan Autonomous Region and identified a total of 301 Tibetan patients without acute stroke. WML severity was graded by the Fazekas Scale. We assessed the overall and age-specific prevalence of WMLs and analyzed associations between WMLs and related factors with univariate and multivariate methods.
Results—
Of the 301 patients, 87 (28.9%) had peripheral vertigo, 83 (27.3%) had primary headache, 52 (17.3%) had a history of stroke, 36 (12.0%) had an anxiety disorder, 29 (9.6%) had epilepsy, 12 (4.0%) had infections of the central nervous system, and 3 (1.0%) had undetermined diseases. WMLs were present in 245 (81.4%) patients, and 54 (17.9%) were younger than 40 years. Univariate analysis showed that age, history of cerebral infarction, hypertension, the thickness of the common carotid artery intima, and plaque within the intracarotid artery were related risks for WMLs. Ordered logistic analysis showed that age, history of cerebral ischemic stroke, hypertension, male sex, and atrial fibrillation were associated with WML severity.
Conclusions—
Risk factors for WMLs appear similar for Tibetans residing at high altitudes and individuals living in the plains. Further investigations are needed to determine whether Tibetans residing at high altitudes have a higher burden of WMLs than inhabitants of the plains.
Graph Neural Networks (GNNs) have achieved state-of-the-art performance in many graph data analysis tasks. However, they still suffer from two limitations for graph representation learning. First, they exploit non-smoothing node features which may result in suboptimal embedding and degenerated performance for graph classification. Second, they only exploit neighbor information but ignore global topological knowledge. Aiming to overcome these limitations simultaneously, in this paper, we propose a novel, flexible, and end-to-end framework, Graph Smoothing Splines Neural Networks (GSSNN), for graph classification. By exploiting the smoothing splines, which are widely used to learn smoothing fitting function in regression, we develop an effective feature smoothing and enhancement module Scaled Smoothing Splines (S3) to learn graph embedding. To integrate global topological information, we design a novel scoring module, which exploits closeness, degree, as well as self-attention values, to select important node features as knots for smoothing splines. These knots can be potentially used for interpreting classification results. In extensive experiments on biological and social datasets, we demonstrate that our model achieves state-of-the-arts and GSSNN is superior in learning more robust graph representations. Furthermore, we show that S3 module is easily plugged into existing GNNs to improve their performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.