斑毛沟尾矿库安全运行实践

镇沅金矿斑毛沟尾矿库投入使用后,存在排洪设施损坏、子坝上升速率快、微细粒尾矿沉积慢、滩长短、泥面缓、尾矿堆筑子坝困难等问题.通过采取排洪设施加固和岸坡防护、土石料筑子坝、槽孔管排渗、旋流器分级沉砂筑坝等多种有效措施,确保了尾矿库的安全运行.该研究对微细粒尾矿排放具有示范作用,是一种可借鉴推广的尾矿库安全运行模式.     

采用细胞神经网络快速校正生物芯片图像倾斜

倾斜校正是全自动生物芯片图像处理必不可少的环节。针对以往倾斜校正算法耗时过长,本文提出一种快速倾斜校正算法。算法首先用细胞神经网络（CNN）寻求各个样点的中心,然后进行Hough变换,再计算方差来寻求倾斜角,最后利用CNN灰度图像旋转模板进行倾斜校正。本算法利用了细胞神经网络并行处理的特性,并充分考虑了生物芯片图像的特点。理论分析与试验结果显示本文算法能够准确高速地完成倾斜校正。     

Role of GDF15/MAPK14 Axis in Chondrocyte Senescence as a Novel Senomorphic Agent in Osteoarthritis

Osteoarthritis (OA) is most prevalent in older individuals and exerts a heavy social and economic burden. However, an effective and noninvasive approach to OA treatment is currently not available. Chondrocyte senescence has recently been proposed as a key pathogenic mechanism in the etiology of OA. Furthermore, senescent chondrocytes (SnCCs) can release various proinflammatory cytokines, proteolytic enzymes, and other substances known as the senescence-associated secretory phenotype (SASP), allowing them to connect with surrounding cells and induce senesce. Studies have shown that the pharmacological elimination of SnCCs slows the progression of OA and promotes regeneration. Growth differentiation factor 15 (GDF15), a member of the tumor growth factor (TGF) superfamily, has recently been identified as a possible aging biomarker and has been linked to a variety of clinical conditions, including coronary artery disease, diabetes, and multiple cancer types. Thus, we obtained data from a publicly available single-cell sequencing RNA database and observed that GDF15, a critical protein in cellular senescence, is highly expressed in early OA. In addition, GDF15 is implicated in the senescence and modulation of MAPK14 in OA. Tissue and synovial fluid samples obtained from OA patients showed overexpression of GDF15. Next, we treated C20A4 cell lines with interleukin (IL)-1β with or without shGDF15 then removed the conditioned medium, and cultured C20A4 and HUVEC cell lines with the aforementioned media. We observed that C20A4 cells treated with IL-1β exhibited increased GDF15 secretion and that chondrocytes cultured with media derived from IL-1β–treated C20A4 exhibited senescence. HUVEC cell migration and tube formation were enhanced after culturing with IL-1β-treated chondrocyte media; however, decreased HUVEC cell migration and tube formation were noted in HUVEC cells cultured with GDF15-loss media. We tested the potential of inhibiting GDF15 by using a GDF15 neutralizing antibody, GDF15-nAb. GDF15-nAb exerted a similar effect, resulting in the molecular silencing of GDF15 in vivo and in vitro. Our results reveal that GDF15 is a driver of SnCCs and can contribute to OA progression by inducing angiogenesis.     

池塘工业化养殖与传统池塘养殖模式对大口黑鲈肌肉品质特性的比较研究

本文旨在比较两种养殖模式下大口黑鲈肌肉的品质差异,以期为池塘工业化养殖模式的改进与推广提供有价值的理论指导。选取2月龄幼鱼(体长约5 cm)随机分为静水养殖组和流水养殖组,分别采用传统池塘养殖和流水槽养殖(流速为3 cm/s)。培养12个月后,采集各组鱼体背部轴上肌,采用常规方法对鱼肉营养成分组成及质构特性进行测定,利用气相色谱-质谱联用仪(GC-MS)和高效液相色谱(HPLC)对肌肉脂肪酸、氨基酸进行测定。结果显示,与静水养殖组比较,流水养殖的鲈鱼肌肉蛋白质含量显著提高(p<0.05),羟脯氨酸、胶原和胶原蛋白含量极显著提高(p<0.01),脂肪含量显著降低(p<0.05),流水养殖组肌肉的质构分析显示硬度、弹性、胶黏性和咀嚼性均显著提高(p<0.05)。此外,流水养殖显著提高了鱼肉多不饱和脂肪酸(PUFA)、ω-6脂肪酸以及油酸、亚油酸、二十碳五烯酸(EPA)和二十二碳六烯酸(DHA)含量。同时,必需氨基酸含量,尤其是谷氨酸(Glu)、缬氨酸(Val)、亮氨酸(Leu)、异亮氨酸(Ile)含量均有显著性提高(p<0.05)。结果表明,相较于传统池塘养殖,池塘工业化养殖模式可提升鲈鱼营养价值,使肉质更加紧实富有嚼劲。     

胞腔同调边缘学习算法研究

边缘划分问题是数据分析的核心问题之一.针对晶体数据的边缘分类问题,引入同调论的思想,提出了胞腔同调边缘算法和正则胞腔同调边缘学习算法及上同调边缘学习算法,并将其应用于晶体结构预测和分类.鉴于晶体数据满足对称群的基本性质,引用同调代数的方法从机器学习的角度来研究数据的边缘分类问题.为了从不同角度构造分类模型,先从相对同调边缘展开为局部同调和定向同调,再深入到上同调边缘算法和腔胞同调边缘算法,由着重系数定理扩展到正则胞腔同调,进而延伸至相对流形.仿真结果表明了算法的有效性.     

Numerical model of a solar pond

This paper points out an idealization of considerable significance in a recent numerical model of a solar pond due to Wang and Akbarzadeh [3] and outlines the refinements required in the formulation of the above model. Typical temperature calculations from teh resultant model are also presented. of the above model. Typical temperature calculations from the resultant model are also presented.     

利用细胞自动机模型约减最近邻分类规则

目前常见的最近邻分类规则约减算法,只注重约减后分类器的分类精度和被约减的规则数量,而不注重约减效率和约减后分类器的泛化能力。针对该问题,提出了一种细胞自动机(cellular automata,CA)基础上的最近邻分类规则约减方法。该方法只保留不同类边界上的样本点,约减规则的数量可以由细胞自动机网格的粒度动态调节。其优势在于:在给定的大数据集前提下,可以利用较少的运行时间来约减给定的规则样本;可以利用积累或迭代的方式来分步获得原给定样本集的一致性子集。采用13个不同的数据集进行仿真实验,结果显示该算法简单、有效,较好地解决了大样本集的约减问题。     

快速发展的蜂窝网通信和卫星通信

主要说明无线通信将应用于移动通信、个人通信和全球通信。数字蜂窝网将增设微蜂窝区 ,频率从 90 0MHz升至 1 .8～ 2GHz ,向宽带和多媒体方向发展 ,数据速率从 1 44kb /s升至 2Mb /s。手机将是TDMA /CDMA双模。蜂窝网不仅用于移动通信 ,而且用于固定用户的市内接入网。卫星通信将加快发展低轨道 /中轨道卫星通信网 ,而同步卫星仍需利用。移动卫星网可通电话 ,可进行多媒体通信 ,速率为 2Mb /s,并通过地面信关站 ,与陆地固定通信网结合一起 ,实现全球个人通信。     

基于细胞自动机的城市用地仿真研究

随着城市规模日益增大,功能日益复杂,城市问题的广度和深度是人类社会前所未遇的,其中土地利用是城市问题的重要方面。从城市交通与土地利用互馈关系入手,应用细胞自动机模型对城市土地利用发展情况进行分析研究、预测和模拟演示,以期能对土地管理提供较为科学的决策依据。     

结合灰度直方图和细胞自动机的多模态MRI脑胶质瘤分割

为了解决脑胶质瘤边界模糊、复杂而导致的分割不准确问题,提出了一种将灰度直方图（gray level histogram,GLH）与改进的细胞自动机相结合的脑胶质瘤分割算法。首先,对脑胶质瘤的T2加权图像和液体衰减反转（FLAIR）图像进行融合;然后,利用灰度直方图特性增强脑胶质瘤区域;最后,以加权距离为特征向量用改进的细胞自动机进行分割,并得到脑胶质瘤各组织分割结果。在20组BraTS2015（brain tumor segmentation）数据库数据和10组临床脑胶质瘤数据上进行分割实验,整个肿瘤区域及核心肿瘤区域的平均分割准确率分别达到90.76%和89.73%。实验结果表明,相对于对比方法,所提算法不仅能更好地分割出对比度明显的胶质瘤区域,还在一定程度上解决了模糊胶质瘤区域分割不准确的问题。该算法在保持不增加算法复杂度的同时,亦提高了算法分割的准确性和鲁棒性。