No abstract

Accumulation of unfolded proteins can disturb the functions of the endoplasmic reticulum (ER), leading to ER-stress or unfolded protein response (UPR). Recent data have shown that activation of UPR can be found in postmortem brains of Alzheimer's disease (AD) patients; and biological markers for activation of UPR are abundant in neurons with diffuse phosphorylated tau. Although these observations suggest a linkage between ER-stress and tau pathology, little is known of their relationship. In this study, we found that high levels of phosphorylated PKR-like ER-resident kinase (p-PERK) and phosphorylated eukaryotic initiation factor 2 alpha (p-eIF2α) as markers for activation of UPR in the hippocampus of aged P301L mutant tau transgenic mice. The immunoreactivity of p-PERK was found to co-localize with that of phosphorylated tau. We then hypothesized that phosphorylation of tau could induce ER-stress and vice versa in promoting AD-like pathogenesis. By using the protein phosphatase 2A inhibitor okadaic acid (OA) as an inducer for phosphorylation of tau, we found that primary cultures of rat cortical neurons treated with OA triggered UPR as indicated by increased levels of p-PERK and p-eIF2α, splicing of mRNA for xbp-1 and elevated levels of mRNA for GADD153. On the other hand, thapsigargin as an ER-stress inducer stimulated phosphorylation of tau at Thr231, Ser262 and Ser396. Thapsigargin also induced activation of caspase-3 and cleavage of tau. These findings suggested that ER-stress and hyperphosphorylation of tau could be induced by each other to form a vicious cycle to propagate AD-like neurodegeneration.

P rimary pontine hemorrhage (PPH) is the most devastating type of intracranial hemorrhage (ICH), with an acute mortality ranging from 30% to 60%. [1][2][3] Various factors including coma at admission, location, and volume of the hematoma were found to associate with these diverse outcomes. However, inconsistent predictors were reported when different parameters and populations were brought into analysis. [4][5][6][7] As a corollary, it would make cogent sense by combining significant factors into a grading scale to enhance predictive power and, meanwhile, provide a useful tool for physicians in decision-making when facing such patients. 8 Currently, there is no standard, widely accepted early prognostic model or clinical grading scale for outcome prediction in PPH patients. The ICH score, composed of age, Glasgow Coma Scale (GCS), infratentorial origin, intraventricular hemorrhage, and hemorrhage volume, has been widely validated in predicting acute mortality, as well as long-term functional outcome in spontaneous ICH.9,10 Easy to use though, the ICH score and its derivatives tended to deem infratentorial hemorrhage as an independent predictor of poor outcome. 11,12 Because of the small structure of the pons but severe manifestations caused by hemorrhagic impairment, the cutoff value Background and Purpose-We aimed to develop and validate a grading scale for predicting 30-day mortality and 90-day functional outcome in patients with primary pontine hemorrhage (PPH). Methods-We retrospectively reviewed records of consecutive patients with first-ever pontine hemorrhage from 3 teaching hospitals between 2005 and 2012. Independent factors associated with 30-day mortality were identified by logistic regression to establish a risk stratification scale, named the new PPH score. For validation of the new PPH score, we prospectively recruited subjects from 10 units between December 2014 and November 2015. The performance of the new PPH score was presented as discrimination and calibration, measured by area under the curve of the receiver operating characteristic and Hosmer-Lemeshow goodness-of-fit, respectively. Results-Data of 171 patients were available for scale development. The new PPH score consisted of 2 independent factors with individual points assigned as follows: Glasgow Coma Scale score 3 to 4 (=2 points), 5 to 7 (=1 point), and 8 to 15 (=0 point); PPH volume >10 mL (=2 points), 5 to 10 mL (=1 point), and <5 mL (=0 point). An independent cohort of 98 patients was applied as an external validation of the new PPH score. Results showed that the new PPH score was discriminative in predicting both 30-day mortality (area under the curve, 0.902) and 90-day good outcome (area under the curve, 0.927). Furthermore, the new PPH score revealed a good calibration (χ 2 =1.387; P=0.846) in 30-day mortality prediction. Conclusions-The

No abstract

No abstract

Graph embedding is an effective yet efficient way to convert graph data into a low dimensional space. In recent years, deep learning has applied on graph embedding and shown outstanding performance. Adjacency matrix is often taken as the storage data structure of graph. However, there are the problems of insufficient spatial proximity information in adjacency matrix. Therefore, this study proposes a deep community detection method which includes (1) matrix reconstruction method, (2) spatial feature extraction method and (3) community detection method. The original adjacency matrix in social network is reconstructed based on the opinion leader and nearer neighbors for obtaining spatial proximity matrix. The spatial proximity matrix can obtain subspace of the graph which can help convolution neural network easily and quickly extract the spatial localization. The spatial eigenvector of reconstructed adjacency matrix can be extracted by an auto-encoder based on convolution neural network for the improvement of modularity. In experiments, four open datasets of practical social networks were selected to evaluate the proposed method, and the experimental results show that the proposed deep community detection method obtained higher modularity than other deep learning methods. Therefore, the proposed deep community detection method can effectively detect high quality communities in social networks.INDEX TERMS Community detection, social network, convolutional neural network, auto-encoder.

This study proposes a mobile positioning method that adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular networks and Wi-Fi networks) for estimating the locations of mobile stations. The recurrent neural networks with multiple consecutive timestamps can be applied to extract the features of time series data for the improvement of location estimation. In practical experimental environments, there are 4525 records, 59 different base stations, and 582 different Wi-Fi access points detected in Fuzhou University in China. The lower location errors can be obtained by the recurrent neural networks with multiple consecutive timestamps (e.g., two timestamps and three timestamps); from the experimental results, it can be observed that the average error of location estimation was 9.19 m by the proposed mobile positioning method with two timestamps.

No abstract