High levels of LDL-C combined with low levels of HDL-C further increase platelet activation in hypercholesterolemic patients

High levels of low-density lipoprotein cholesterol (LDL-C) enhance platelet activation, whereas high levels of high-density lipoprotein cholesterol (HDL-C) exert a cardioprotective effect. However, the effects on platelet activation of high levels of LDL-C combined with low levels of HDL-C (HLC) have not yet been reported. We aimed to evaluate the platelet activation marker of HLC patients and investigate the antiplatelet effect of atorvastatin on this population. Forty-eight patients with high levels of LDL-C were enrolled. Among these, 23 had HLC and the other 25 had high levels of LDL-C combined with normal levels of HDL-C (HNC). A total of 35 normocholesterolemic (NOMC) volunteers were included as controls. Whole blood flow cytometry and platelet aggregation measurements were performed on all participants to detect the following platelet activation markers: CD62p (P-selectin), PAC-1 (GPIIb/IIIa), and maximal platelet aggregation (MPAG). A daily dose of 20 mg atorvastatin was administered to patients with high levels of LDL-C, and the above assessments were obtained at baseline and after 1 and 2 months of treatment. The expression of platelets CD62p and PAC-1 was increased in HNC patients compared to NOMC volunteers (P<0.01 and P<0.05). Furthermore, the surface expression of platelets CD62p and PAC-1 was greater among HLC patients than among HNC patients (P<0.01 and P<0.05). Although the expression of CD62p and PAC-1 decreased significantly after atorvastatin treatment, it remained higher in the HLC group than in the HNC group (P<0.05 and P=0.116). The reduction of HDL-C further increased platelet activation in patients with high levels of LDL-C. Platelet activation remained higher among HLC patients regardless of atorvastatin treatment.

Prenatal diagnosis by array-comparative genomic hybridization.

BACKGROUND: Array-comparative genomic hybridization (aCGH) was developed as a high-resolution analysis of DNA copy number variations, initially in cancer studies, and subsequently extended to postnatal evaluation of mental retardation and multiple congenital anomalies. OBJECTIVE: To review the current and potential applications of aCGH in prenatal diagnosis. METHODS: The role of aCGH is compared with conventional prenatal diagnostic methods regarding turnaround time, resolution and cost. The challenges and concerns are also discussed. RESULTS/CONCLUSIONS: Array-CGH offers a rapid analysis of the DNA copy number variations with results of a comprehensive genome-wide picture available within 3 days' time, much shorter than the G-banded analysis, which takes at least 2 weeks. In addition, its superior resolution allows detection of submicroscopic microdeletions or microduplications, and a more precise delineation of chromosomal aberrations, leading to improved genotype-phenotype correlations. However, aCGH cannot detect truly balanced chromosomal rearrangements or polypoidy, and may even generate data with unknown significance. Knowing its limitations and with proper counseling of the advantages and shortcomings of aCGH, the authors believe aCGH will become the first-line diagnostic test for management of pregnancy with fetal sonographic anomalies.

Extended Kalman Filter-Based Pruning Method for Recurrent Neural Networks.

Pruning is one of the effective techniques for improving the generalization error of neural networks. Existing pruning techniques are derived mainly from the viewpoint of energy minimization, which is commonly used in gradient-based learning methods. In recurrent networks, extended Kalman filter (EKF)-based training has been shown to be superior to gradient-based learning methods in terms of speed. This article explains a pruning procedure for recurrent neural networks using EKF training. The sensitivity of a posterior probability is used as a measure of the importance of a weight instead of error sensitivity since posterior probability density is readily obtained from this training method. The pruning procedure is tested using three problems: (1) the prediction of a simple linear time series, (2) the identification of a nonlinear system, and (3) the prediction of an exchange-rate time series. Simulation results demonstrate that the proposed pruning method is able to reduce the number of parameters and improve the generalization ability of a recurrent network.