No abstract

Extracellular vesicles (EVs) derived from mesenchymal stem cells (MSCs) have emerged as important mediators of intercellular communication in response to cartilage damage. In this study, we sought to characterize the inhibitory role of microRNA (miR)-31 encapsulated in synovial MSC (SMSC)-derived EVs in knee osteoarthritis (OA). The expression of miR-31, lysine demethylase 2A (KDM2A), E2F transcription factor 1 (E2F1), and pituitary tumor transforming gene 1 (PTTG1) was validated in cartilage tissues of knee OA patients. Following SMSC-EV extraction and identification, chondrocytes with the miR-31 inhibitor were added with SMSC-EVs, whereupon the effects of miR-31 on proliferation and migration of chondrocytes were assessed. The interaction among miR-31, KDM2A, E2F1, and PTTG1 in chondrocyte activities was probed in vitro , along with an in vivo mouse knee OA model. We identified downregulated miR-31, E2F1, and PTTG1 and upregulated KDM2A in cartilage tissues of knee OA patients. SMSC-EV-packaged miR-31 potentiated chondrocyte proliferation and migration as well as cartilage formation by targeting KDM2A. Mechanistically, KDM2A bound to the transcription factor E2F1 and inhibited its transcriptional activity. Enrichment of E2F1 in the PTTG1 promoter region activated PTTG1 transcription, accelerating chondrocyte proliferation and migration. SMSC-EVs and EVs from miR-31-overexpressed SMSCs alleviated cartilage damage and inflammation in knee joints in vivo . SMSC-EV-encapsulated miR-31 ameliorates knee OA via the KDM2A/E2F1/PTTG1 axis.

No abstract

No abstract

Abstract:The dynamic window approach has the drawback that it may result in local minima and nonoptimal motion decision for obstacle avoidance because of not considering the size constraint of a mobile robot. Thus, an improved dynamic window approach is proposed, which takes into account the relation between the size of the mobile robot and the free space between obstacles. A laser range finder is employed to improve the ability of sensing and prediction of the environment, in order to avoid being trapped in a U-shaped obstacle, such as a box canyon. By applying the proposed method, the local minima problem can be solved and the optimal path can be obtained. The effectiveness and superiority is proven by theoretic analysis and simulations.

No abstract

No abstract

Deep Neural Networks (DNNs) have revolutionized a wide range of industries, from healthcare and finance to automotive, by offering unparalleled capabilities in data analysis and decisionmaking. Despite their transforming impact, DNNs face two critical challenges: the vulnerability to adversarial attacks and the increasing computational costs associated with more complex and larger models. In this paper, we introduce an effective method designed to simultaneously enhance adversarial robustness and execution efficiency. Unlike prior studies that enhance robustness via uniformly injecting noise, we introduce a non-uniform noise injection algorithm, strategically applied at each DNN layer to disrupt adversarial perturbations introduced in attacks. By employing approximation techniques, our approach identifies and protects essential neurons while strategically introducing noise into non-essential neurons. Our experimental results demonstrate that our method successfully enhances both robustness and efficiency across several attack scenarios, model architectures, and datasets.