摘要：The aim of this study was to evaluate the efficacy of mosaicplasty with tissue-engineered cartilage for the treatment of osteochondral defects in a pig model with advanced MR technique. Eight adolescent miniature pigs were used. The right knee underwent mosaicplasty with tissue-engineered cartilage for treatment of focal osteochondral defects, while the left knee was repaired via single mosaicplasty as controls. At 6, 12, 18 and 26 weeks after surgery, repair tissue was evaluated by magnetic resonance imaging （MRI） with the cartilage repair tissue （MOCART） scoring system and T2 mapping. Then, the results of MRI for 26 weeks were compared with findings of macroscopic and histologic studies. The MOCART scores showed that the repaired tissue of the tissue-engineered cartilage group was statistically better than that of controls （P 〈 0.001）. A significant correlation was found between macroscopic and MOCART scores （P 〈 0.001）. Comparable mean T2 values were found between adjacent cartilage and repair tissue in the experimental group （P 〉 0.05）. For zonal T2 value evaluation, there were no significant zonal T2 differences for repair tissue in controls （P 〉 0.05）. For the experimental group, zonal T2 variation was found in repair tissue （P 〈 0.05）. MRI, macroscopy and histology showed better repair results and bony incorporation in mosaicplasty with the tissue-engi- neered cartilage group than those of the single mosaicplasty group. Mosaicplasty with the tissue-engineered cartilage is a promising approach to repair osteochodndral defects. Morphological MRI and T2 mapping provide a non-invasive method for monitoring the maturation and integration of cartilage repair tissue in vivo.

摘要：[目的/意义]对人工智能领域科研团队进行识别,并基于多个维度的指标提取领军科研团队,旨在丰富科研团队识别的流程与方法,为从科研团队视角分析人工智能领域脉络、前沿和主题提供依据。[方法/过程]以Web of Science为数据来源,采集2009-2018年间人工智能学科领域所有科技论文的数据,通过算法设计与人工核查进行数据清洗;基于分数计数法构建全局合著网络,并利用社区探测算法动态调参、识别科研团队;进而基于多维度的指标提取出领军团队,并加以比较分析。[结果/结论]从实践出发构造人工智能科技论文数据清洗的规则;构建基于合著关系识别人工智能科研团队的流程体系;提出通过消除边缘结点进行合著网络筛选,进而利用已知团队作为参考进行参数调整的思路;较为系统和准确地识别出全球人工智能科研团队,并基于发文量、被引量、h指数、中介中心度、接近中心度和加权点度中心度6个维度的指标提取出领军科研团队,同时,给出结合论文数据和实证调研对每个领军团队的示例性分析。

摘要：设计了一种基于模糊PID控制器和超磁致伸缩致动器的车削振动主动控制系统,推导了超磁致伸缩致动器和专用刀架的数学模型,并在Matlab环境下进行建模仿真,当振动频率在50~100Hz时,仿真结果表明该控制系统能抑制振幅达50%以上;在现场车削试验中,以普通45#钢为车削工件,分别在转速1 500r/min、车削深度为0.07mm和转速3 000r/min,车削深度为0.04mm下,对车削振动进行主动控制,结果证明所设计的车削振动控制系统能够抑制振动幅度达30%以上。

摘要：[目的/意义]随着数据生态环境越来越复杂多变,大量真实而有价值的数据常常隐藏在海量多源异构数据中。如何高效地从多源情报信息中采集与整合数据并有效支撑深入分析,一直是数据科学领域的核心问题,也是情报学领域的重要基础工作。[方法/过程]首先,从多个角度对多源情报信息环境进行解析,分析不同视角下信息环境的主要特征。其次,融合Hook机制进行信息采集方法研究,探索Hook作用于信息传递过程的机理,并与Python采集框架相结合。然后,设计与构建多源情报信息采集框架,确定框架的实施方式与作用范围,并扩展采集框架至知识与智慧层次,以实现价值转换。最后,针对多源情报信息类型应用本框架实施采集。[结果/结论]在面对复杂的多源情报信息时,本研究提出的多源情报信息采集框架能够准确有效地获取数据,并通过元数据整合为后续知识挖掘研究奠定了扎实的数据基础。