基于模糊自适应共振网络简化模型方法的乳腺肿瘤基因分类

以乳腺肿瘤的基因微阵列实验所产生的基因微阵列表达数据为研究对象,针对大规模基因表达模式分析的特点,提出基于模糊自适应共振网络简化模型的基因表达数据分类方法,并在此基础上实现了BRCA1变异和BRCA2变异的乳腺肿瘤分类.利用Hedenfalk的基因微阵列数据,对BRCA1变异和BRCA2变异作分类计算,计算结果达到78%以上的正确率.     

Serum microRNA-155 as a potential biomarker for breast cancer screening

MicroRNAs(miRs) have been shown to be differentially expressed in the serum of cancer patients and controls,and can thus be used as biomarkers for cancer screening.We detected the expression level of miR-155 in the serum of female breast cancer patients and healthy controls to investigate whether serum miR-155 could discriminate patients with early-stage breast cancer.Serum samples were collected from 20 female patients with newly diagnosed breast cancer and 10 healthy controls.Real-time quantitative PCR was used to detect the expression level of miR-155.The expression level of miR-155 was significantly increased in the serum of breast cancer patients compared with in the serum of normal controls.MiR-155 may be useful as a blood-based biomarker for breast cancer screening.     

人类MMP3基因2号外显子Lys45Glu单核苷酸多态性与乳腺癌易感性的关系

讨论了基质金属蛋白酶3(MMP3)基因单核苷酸多态性(SNP)与湖北省武汉市人群乳腺癌易感性的关联关系.选择MMP3基因2号外显子Lys45Glu单核苷酸多态性rs679620作为研究对象,采用聚合酶链式反应限制性片段长度多态性分析方法(PCR-RFLP)对195例散发乳腺癌病人和289例正常人群进行了病例-对照分析,运用SHEsis软件和SPSS软件分别计算Hardy-Weinberg数值以及对基因分型数据进行统计分析后表明:MMP3 rs679620(A/G)单核苷酸多态性各个基因型在病例组和对照组分布符合Hardy-Weinberg平衡,A等位基因在病例组中频率(P=0.331)小于对照组(P=0.360),可能为保护因素,但病例和对照的差异不具有统计学意义.在中国武汉地区汉族女性中,MMP3多态性与乳腺癌没有明显关联,但呈现一定的趋势,需加大样本量和增加考察的SNP数量进行深入研究.     

回收废弃印刷电路板焊锡的新技术

提出一种回收废弃印刷电路板焊锡的有效方法.其步骤为:在整个回收过程中不会对环境造成二次污染:用油加热废弃印刷电路板至焊锡熔化,在离心力的作用下将焊锡分离和回收.实验结果表明:当油温为240℃,转鼓转速为1 400r/min,采用间歇式的方式旋转6min,废弃印刷电路板焊锡即可回收完全.     

Monoclonal antibodies against antigens on breast cancer cells

Summary Of 360 mAb obtained in a cell fusion experiment with the spleen cells of a mouse immunized with a mixture of different human breast carcinoma cells lines, 30 mAb were selected which reacted more strongly with tumor cells than with (noncancerous) fibroblasts. Theses mAb were tested for reactivity with additional types of cancerous and noncancerous tissues. Two mAb showed high tumor selectivity, but the corresponding epitopes on individual tumor cells were heterogeneously expressed. The mAb will be evaluated for in vivo applications.     

关于内分泌激素的Logistic建模分析

随着Logistic回归分析模型近年在医学资料分析中逐渐得到认可,本文采用Logistic回归方法对乳腺肿瘤病与六种内分泌激素的病理学关系进行了分析,建立了该病与相关变量的关系模型,分析出在其它自变量不变时,有统计意义的某自变量任两水平间的发生比与对应结果y取任何值的概率间大小关系,并作出判断相关激素与乳腺疾病关系的激素范围,取得了符合病理学理论和临床经验的结果,对相关激素与乳腺疾病关系的研究、临床诊断和教学有参考价值.     

The application of the edge sharpening operator to the breast near-infrared imaging

The principles of Gradient operator, Laplacian operator, LOG operator and Sobel operator are discussed. Certain features of breast can be extracted in different degrees and aspects from original images by applying four edge sharpening operators to the breast near-infrared imaging. A great number of cases prove that compared with the other three operators, the improved Sobel operator can effectively extract the structural features of the breast from an original image. It can be concluded that the improved Sobel operator can assist in diagnosing breast diseases. Foundation item: Supported by the Enterprise Innovation Foundation of Wuhan Biography: Xiang Yan (1979-), female, Master candidate, research direction: biomedical engineering and biophysics.     

溴化环氧树脂印刷线路板的中温热解实验

利用热重法进行了溴化环氧树脂(FR4)印刷线路板不同升温速率的热解特性试验研究,并利用热重和热重微分曲线的信息计算了动力学三参数.利用固定床热解试验装置进行了大颗粒FR4型线路板中温热解实验,并利用GC-MS,FT-IR和SEM检测了热解产物.结果表明:①FR4型线路板热解温度区间为300,℃～400,℃,热解反应活化能107.70 kJ/mol,指前因子为2.22×109/s,反应级数为1;②热解气体产物主要为CO、CO2、溴甲烷及少量的丙酮和C1～C3卤代烷烃;热解液体产物主要为苯酚和烷基苯酚,还有其他C6～C17芳香族如苯、烷基苯等成分;③固体产物包括金属片和含碳的玻璃纤维,可手动实现金属和玻璃纤维的分离富集,富集的金属包含大量的铜和少量的金、镍等贵重金属,富集的玻璃纤维包含碳、硅、氧、钙和铝等元素.     

Review：BRCA1/2 associated hereditary breast cancer

Breast cancer is one of the leading causes of death in women today. Some of the patients are hereditary, with a large proportion characterized by mutation in BRCAI and/or BRCA2 genes. In this review, we provide an overview of these two genes, focusing on their relationship with hereditary breast cancers. BRCA1/2 associated hereditary breast cancers have unique features that differ from the general breast cancers, including alterations in cellular molecules, pathological bases, biological behavior, and a different prevention strategy. But the outcome of BRCA1/2 associated hereditary breast cancers still remains controversial; further studies are needed to elucidate the nature of BRCA 1/2 associated hereditary breast cancers.     

基于支持向量机的乳腺癌辅助诊断

采用支持向量机、K-近邻法(K-Nearest Neighbor,K-NN)、概率神经网络(Probabilistic Neural Network,PNN),结合乳腺肿瘤的细针穿刺细胞病理学临床数据诊断乳腺癌.结果表明:当使用sigmoid核函数时,SVM通过5次交叉验证的最佳平均分类准确率达到了96.24%,优于K-NN(95.37%),PNN(95.09%)等分类器,表明该方法有望成为一种实用的乳腺癌临床辅助诊断工具.