Time-Series-Based Personalized Lane-Changing Decision-Making Model.

In recent years, autonomous driving technology has been changing from "human adapting to vehicle" to "vehicle adapting to human". To improve the adaptability of autonomous driving systems to human drivers, a time-series-based personalized lane change decision (LCD) model is proposed. Firstly, according to the characteristics of the subject vehicle (SV) with respect to speed, acceleration and headway, an unsupervised clustering algorithm, namely, a Gaussian mixture model (GMM), is used to identify its three different driving styles. Secondly, considering the interaction between the SV and the surrounding vehicles, the lane change (LC) gain value is produced by developing a gain function to characterize their interaction. On the basis of the recognition of the driving style, this gain value and LC feature parameters are employed as model inputs to develop a personalized LCD model on the basis of a long short-term memory (LSTM) recurrent neural network model (RNN). The proposed method is tested using the US Open Driving Dataset NGSIM. The results show that the accuracy, F1 score, and macro-average area under the curve (macro-AUC) value of the proposed method for LC behavior prediction are 0.965, 0.951 and 0.983, respectively, and the performance is significantly better than that of other mainstream models. At the same time, the method is able to capture the LCD behavior of different human drivers, enabling personalized driving.

[6-Gingerol Pretreatment Alleviates Hypoxia/Reoxygenation-induced H9C2 Cardiomyocyte Injury by Inhibiting Oxidative Stress and Inflammation].

OBJETIVE: To observe the effect of 6-gingerol (6-G) pretreatment on hypoxia/reoxygenation (H/R) induced injury in H9C2 myocardial cell and investigate its related mechanism. METHODS: The H/R in vitro model of cardiomyocytes was prepared by conventional methods. In detail, H9C2 cells were added with the nitrogen-saturated hypoxic liquid, and placed in an incubator, mixed with gas (1% O 2, 5% CO 2, 94% N 2) applying for 15 min. After culturing for 3 h, the cells were taken out and placed in an incubator (37℃, 5% CO 2) for 1 h. Before establishing the cell model, the cells were pretreated with 6-G, and the cell viability was measured by MTT method to observe the protective effect of different concentrations of 6-G on H/R-induced cell damage. The 6-G mass concentration for pretreatment that led to the highest cell viability was used for follow-up experiments. DCFH-DA fluorescent probe was used to detect the effect of 6-G pretreatment on H9C2 oxidative stress level, and the intracellular oxidative stress was observed with fluorescence microscope and flow cytometry. Western blot method was used to detect the expression of tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6) and interleukin-1ß (IL-1ß) in H/R-induced cell inflammatory responses. RESULTS: Compared with the H/R group, the cell viability of the 6-G+H/R group began to increase when the concentration of 6-G promoted to50 µg/mL. The cell viability was the highest after pretreated with 200 µg/mL 6-G. Therefore, 200 µg/mL was considered as the best 6-G intervention concentration for subsequent experiment. The content of reactive oxygen species (ROS) in the 200 µg/mL 6-G group had no significant changes compared with the control group (P>0.05), and the ROS fluorescence peak did not migrate significantly. However the ROS content in the H/R group increased significantly compared with the control (P<0.05), and the ROS fluorescence peak shifted to the right. Compared with the H/R group, the ROS content of the 6-G+H/R group decreased (P<0.05), and the ROS fluorescence peak shifted to the left. Compared with the control group, the expressions of TNF-α, IL-6, IL-1ß in the 6-G group had no significant changes (P>0.05); the expressions of TNF-α, IL-6, IL-1ß in the H/R group increased (P<0.05). Compared with H/R group, the expressions of TNF-α, IL-6 and IL-1ß in 6-G+H/R group decreased (P<0.05). CONCLUSION: 6-G pretreatment can alleviate H/R-induced H9C2 myocardial injury, which may be related to the inhibition of oxidative stress and inflammatory responses.

[Advances in the relationship between leptin and hypertensive-left ventricular hypertrophy].

Leptin is a protein hormone produced mainly by obese gene and secreted by adipose tissue and exerts the biological effects through leptin receptors. With the progress in research on the function and receptor signal transduction related leptin and leptin resistance, it has been found that leptin is associated with the development and progression of many cardiovascular diseases, such as hypertension and left ventricular hypertrophy. Some studies have reported that leptin resistance is the pathologic basis for a variety of cardiovascular diseases. This paper will briefly review the advances in the study of correlation between leptin and hypertensive-left ventricular hypertrophy (HLVH), focusing on the relationship between leptin and various factors related to HLVH, such as sympathetic nervous system, renin angiotensin aldosterone system, growth factors, inflammatory factors and insulin resistance.

From Global to Local: Multi-Patch and Multi-Scale Contrastive Similarity Learning for Unsupervised Defocus Blur Detection.

Defocus blur detection (DBD), which aims to detect out-of-focus or in-focus pixels from a single image, has been widely applied to many vision tasks. To remove the limitation on the abundant pixel-level manual annotations, unsupervised DBD has attracted much attention in recent years. In this paper, a novel deep network named Multi-patch and Multi-scale Contrastive Similarity (M2CS) learning is proposed for unsupervised DBD. Specifically, the predicted DBD mask from a generator is first exploited to re-generate two composite images by transporting the estimated clear and unclear areas from the source image to realistic full-clear and full-blurred images, respectively. To encourage these two composite images to be completely in-focus or out-of-focus, a global similarity discriminator is exploited to measure the similarity of each pair in a contrastive way, through which each two positive samples (two clear images or two blurred images) are enforced to be close while each two negative samples (a clear image and a blurred image) are inversely far. Since the global similarity discriminator only focuses on the blur-level of a whole image and there do exist some fail-detected pixels which only cover a small part of areas, a set of local similarity discriminators are further designed to measure the similarity of image patches in multiple scales. Thanks to this joint global and local strategy, as well as the contrastive similarity learning, the two composite images are more efficiently moved to be all-clear or all-blurred. Experimental results on real-world datasets substantiate the superiority of our proposed method both in quantification and visualization. The source code is released at: https://github.com/jerysaw/M2CS.