Two-timescale projection neural networks in collaborative neurodynamic approaches to global optimization and distributed optimization.

In this paper, we propose a two-timescale projection neural network (PNN) for solving optimization problems with nonconvex functions. We prove the convergence of the PNN with sufficiently different timescales to a local optimal solution. We develop a collaborative neurodynamic approach with multiple such PNNs to search for global optimal solutions. In addition, we develop a collaborative neurodynamic approach with multiple PNNs connected via a directed graph for distributed global optimization. We elaborate on four numerical examples to illustrate the characteristics of the approaches.

To establish a model for the prediction of initial standard and maintenance doses of warfarin for the Han Chinese population based on gene polymorphism: a multicenter study.

PURPOSE: The purpose of this paper is to study the correlation between demographic and clinical factors and warfarin dose of patients in Chinese Han population taking warfarin and study gene polymorphisms impact of related gene loci (CYP2C9*3, VKORC1-1639G > A) on warfarin doses, to establish a model to predict initial standard dose and maintenance dose based on CYP2C9*3, VKORC1-1639G > A genotype. METHODS: The study collects the data of patients in our hospital and other subcenters which incorporates 2160 patients to establish the initial dose model and 1698 patients for the stable dose model, and sequences 26 multigene sites in 451 patients. Based on the patient's dosage, clinical data, and demographic characteristics, the genetic and non-genetic effects on the initial dose and stable dose of warfarin are calculated by using statistical methods, and the prediction model of initial standard dose and maintenance dose can be established via multiple linear regression. RESULTS: The initial dose of warfarin (mg/day) was calculated as (1.346 + 0.350 × (VKORC1-1639G > A) - 0.273 × (CYP2C9*3) + 0.245 × (body surface area) - 0.003 × (age) - 0.036 × (amine-iodine) + 0.021 × (sex))2. This model incorporated seven factors and explained 55.3% of the individualization differences of the warfarin drug dose. The maintenance dose of warfarin (mg/day) was calculated as (1.336 + 0.299 × (VKORC1-1639G > A) + 0.480 × (body surface area) - 0.214 × (CYP2C9*3) - 0.074 × (amine-iodine) - 0.003 × (age) - 0.077 × (statins) - 0.002 × (height))2. This model incorporated six factors and explained 42.4% of the individualization differences in the warfarin drug dose. CONCLUSION: The genetic and non-genetic factors affecting warfarin dose in Chinese Han population were studied systematically in this study. The pharmacogenomic dose prediction model constructed in this study can predict anticoagulant efficacy of warfarin and has potential application value in clinical practice.

Safety of the Xuesaitong injection in China: results from a large-scale multicentre post-marketing surveillance study in a real-world setting.

OBJECTIVE: The safety profile of traditional Chinese medicine injections has emerged as the greatest challenge to their clinical application. The authors aimed to perform a post-marketing surveillance study in a real-world setting to evaluate the safety of the Xuesaitong (XST) injection in China. METHODS: This multi-centre, post-marketing, observational study enrolled patients who received XST injections in 42 centres in China between March 2015 and November 2017. Adverse drug reactions (ADRs) and adverse drug events (ADEs) were collected and evaluated in a post-marketing database. Logistic regression analysis was performed to analyse the risk factors for ADRs. RESULTS: A total of 30,008 consecutive patients with a mean age of 62.29 ± 14.58 years were included in this post-marketing study. The incidences of ADEs and ADRs were 0.5% and 0.33%, respectively. The most common clinical manifestations were damage to skin and appendages (47.66%). There were four new kinds of ADEs found in the present monitoring study. The majority of ADRs were type B (62.62%) and occurred within 24 h after XST injection treatment. No severe ADRs were reported in this analysis. Multivariate logistic regression analysis showed that the hospital level (OR = 0.607; 95% CI = 0.407-0.906; p = .0144), hypertension (OR = 1.979; 95% CI, 1.323-2.959; p = .0009) and solvent type (OR = 2.951; 95% CI, 1.608-5.417; p = .0005) were risk factors for ADR occurrence. CONCLUSION: XST injection is well tolerated and has a favourable safety profile for patients in a real-world setting. This post-marketing study provided further evidence of the safety of XST injections for clinical applications.