This paper presents the design, modeling, and experimental validation of CapsuleBot, a novel hybrid aerial-ground bi-copter robot designed for long-endurance and low-noise operations. CapsuleBot combines the maneuverability of a bi-copter in the air with the low power consumption and low noise of a two-wheel self-balancing robot on the ground. To achieve this, we design an innovative mechanical structure named the actuated-wheel-rotor, which uses a servo motor and a brushless motor to function as both a tilting rotor in the air and an actuated wheel on the ground. CapsuleBot is equipped with two actuated-wheel-rotors, enabling it to achieve hybrid aerial-ground propulsion using only four motors, with no additional motors required compared to a bi-copter. Additionally, we develop comprehensive dynamics and control systems for both air and wheel mode, based on the bi-copter model and the two-wheel self-balancing robot model. A prototype of CapsuleBot is constructed, and its performance in terms of low power consumption and low noise is validated through experiments. Challenging tasks demonstrate CapsuleBot's capability to climb steep, fly over cliffs, and traverse rough terrains.

TLRs are essential for sensing the invading pathogens and initiating protective immune responses. However, aberrant activation of TLR-triggered inflammatory innate responses leads to the inflammatory disorders and autoimmune diseases. The molecular mechanisms that fine-tune TLR responses remain to be fully elucidated. Protein tyrosine phosphatase with proline-glutamine-serine-threonine–rich motifs (PTP-PEST) has been shown to be important in cell adhesion, migration, and also T cell and B cell activation. However, the roles of PTP-PEST in TLR-triggered immune response remain unclear. In this study, we report that PTP-PEST expression was upregulated in macrophages by TLR ligands. PTP-PEST inhibited TNF-α, IL-6, and IFN-β production in macrophages triggered by TLR3, TLR4, and TLR9. Overexpression of catalytically inactive mutants of PTP-PEST abolished the inhibitory effects, indicating that PTP-PEST inhibits TLR response in a phosphatase-dependent manner. Accordingly, PTP-PEST knockdown increased TLR3, -4, and -9–triggered proinflammatory cytokine and type I IFN production. PTP-PEST selectively inhibited TLR-induced NF-κB activation, whereas it had no substantial effect on MAPK and IFN regulatory factor 3 activation. Moreover, PTP-PEST directly interacted with IκB kinase β (IKKβ) then inhibited IKKβ phosphorylation at Ser^177/181 and Tyr^188/199, and subsequently suppressed IKKβ activation and kinase activity as well as downstream NF-κB activation, resulting in suppression of the TLR-triggered innate immune response. Thus, PTP-PEST functions as a feedback-negative regulator of TLR-triggered innate immune responses by selectively impairing IKKβ/NF-κB activation.

Acute ischemic stroke (AIS) accounts for more than 80% of the approximately 610,000 new stroke cases worldwide every year. Both ischemia and reperfusion can cause death, damage, and functional changes of affected nerve cells, and these alterations can result in high rates of disability and mortality. Therefore, therapies aimed at increasing neuroprotection and neurorepair would make significant contributions to AIS management. However, with regard to AIS therapies, there is currently a large gap between experimental achievements and practical clinical solutions (EC-GAP-AIS). Here, by integrating curated disease-gene associations and interactome network known to be related to AIS, we investigated the molecular network mechanisms of multi-module structures underlying AIS, which might be relevant to the time frame subtypes of AIS. In addition, the EC-GAP-AIS phenomenon was confirmed and elucidated by the shortest path lengths and the inconsistencies in the molecular functionalities and overlapping pathways between AIS-related genes and drug targets. Furthermore, we identified 23 potential targets (e.g. ADORA3, which is involved in the regulation of cellular reprogramming and the extracellular matrix) and 46 candidate drugs (e.g. felbamate, methylphenobarbital and memantine) that may have value for the treatment of AIS.

Electroenccephalogram (EEG) is well established for assessing the functional state of the brain. During the anesthesia, the brain activity level changed dramatically, so we could get the depth of anesthesia estimation from EEG recording. In previous research, most achievements were in the frequency-domain. In this paper, besides the frequency-domain features extraction, we have introduced the concept of “Entropy” as well as nonlinear feature. The results show that, with the deepening of anesthesia degree, approximate entropy (ApEn), Shannon entropy (SSE) and Lempel— Ziv complexity from EEG signal decrease gradually. Centre frequency towards to low frequency, total power shows a rising trend. Largest Lyapunov index also decreases, while the correlation dimension has no clear trend

Deep reinforcement learning is one of the core technologies, which leads to artificial intelligence. It has been used widely not only in robot, autonomous driving, power station control and human-computer interaction in mobile e-commerce platforms, but also in gambling games. The game software AlphaGo of artificial intelligence defeated Ke Jie in 2017. Since then, the study of artificial intelligence and deep reinforcement is widely concerned. We build a Neural Network model using TensorFlow and Monte Carlo tree search algorithm, so that agents can reach the level of playing chess with human beings in board games. The gobang interface is realized by Python, and the PhoenixGo recurrence is completed. The computation is reduced and the training effect of the agent is improved by optimizing the establishment of gobang strategic value network. Black hand first can achieve a 73% winning rate, and white hand first is 64%, when playing against human players.