DCRL-KG: Distributed Multi-Modal Knowledge Graph Retrieval Platform Based on Collaborative Representation Learning

The knowledge graph with relational abundant information has been widely used as the basic data support for the retrieval platforms.Image and text descriptions added to the knowledge graph enrich the node information,which accounts for the advantage of the multi-modal knowledge graph.In the field of cross-modal retrieval platforms,multi-modal knowledge graphs can help to improve retrieval accuracy and efficiency because of the abundant relational infor-mation provided by knowledge graphs.The representation learning method is sig-nificant to the application of multi-modal knowledge graphs.This paper proposes a distributed collaborative vector retrieval platform(DCRL-KG)using the multi-modal knowledge graph VisualSem as the foundation to achieve efficient and high-precision multimodal data retrieval.Firstly,use distributed technology to classify and store the data in the knowledge graph to improve retrieval efficiency.Secondly,this paper uses BabelNet to expand the knowledge graph through multi-ple filtering processes and increase the diversification of information.Finally,this paper builds a variety of retrieval models to achieve the fusion of retrieval results through linear combination methods to achieve high-precision language retrieval and image retrieval.The paper uses sentence retrieval and image retrieval experi-ments to prove that the platform can optimize the storage structure of the multi-modal knowledge graph and have good performance in multi-modal space.

Virtual population pharmacokinetic using physiologically based pharmacokinetic model for evaluating bioequivalence of oral lacidipine formulations in dogs

The aim of the present study was to investigate virtual population pharmacokinetic using physiologically based pharmacokinetic(PBPK) model for evaluating bioequivalence of oral lacidipine formulations in dogs. The dissolution behaviors of three lacidipine formulations including one commercial product and two self-made amorphous solid dispersions(ASDs)capsules were determined in 0.07% Tween 80 media. A randomized 3-period crossover design in 6 healthy beagle dogs after oral administration of the three formulations at a single dose of 4 mg was conducted. The PBPK modeling was utilized for the virtual bioequivalence study.In vitro dissolution experiment showed that the dissolution behaviors of lacidipine amorphous solid dispersions(ASDs) capsules, which was respectively prepared by HPMC-E5 or Soluplus, as polymer displayed similar curves compared with the reference formulation in 0.07% Tween 80 media. In vivo pharmacokinetics experiments showed that three formulations had comparable maximum plasma drug concentration(Cmax), and the time(Tmax) to reach Cmax of lacidipine tablet, which was prepared by Soluplus, as polymer was slower than other two formulations in consistency with the in vitro dissolution rate. The 90% confidence interval(CI) for the Cmax, AUC0–24 h and AUC0–∞ of the ratio of the test drug to the reference drug exceeded the acceptable bioequivalence(BE) limits(0.80–1.25). However, the 90% CI of the AUC0–24 h, AUC0–∞ and Cmax of the ratio of test to reference drug were within the BE limit,calculated using PBPK modeling when the virtual subjects reached 24 dogs. The results all demonstrated that virtual bioequivalence study can overcome the inequivalence caused by inter-subject variability of the 6 beagle dogs involved in in vivo experiments.

Spinning from Nature:Engineered Preparation and Application of High-Performance Bio-Based Fibers

Many natural fibers are lightweight and display remarkable strength and toughness.These properties originate from the fibers’hierarchical structures,assembled from the molecular to macroscopic scale.The natural spinning systems that produce such fibers are highly energy efficient,inspiring researchers to mimic these processes to realize robust artificial spinning.Significant developments have been achieved in recent years toward the preparation of high-performance bio-based fibers.Beyond excellent mechanical properties,bio-based fibers can be functionalized with a series of new features,thus expanding their sophisticated applications in smart textiles,electronic sensors,and biomedical engineering.Here,recent progress in the construction of bio-based fibers is outlined.Various bioinspired spinning methods,strengthening strategies for mechanically strong fibers,and the diverse applications of these fibers are discussed.Moreover,challenges in reproducing the mechanical performance of natural systems and understanding their dynamic spinning process are presented.Finally,a perspective on the development of biological fibers is given.