Method of demarcating two-dimensional distribution
    2.
    发明授权
    Method of demarcating two-dimensional distribution 失效
    划分二维分布的方法

    公开(公告)号:US5006986A

    公开(公告)日:1991-04-09

    申请号:US279955

    申请日:1988-12-05

    申请人: Takanobu Inoue

    发明人: Takanobu Inoue

    IPC分类号: G01N33/49 G01N15/14

    摘要: One particle group in a sample undergoing measurement is demarcated from another particle group in the sample by drawing a discriminant on a plot of a two-dimensional distribution of the particles. To draw the discriminant, the two-dimensional distribution is divided into a grid pattern of a plurality of areas arranged in a number of columns, the number of particles contained in each area is determined, and an area containing the smallest number of particles is selected from a plurality of areas in a specific column. Next, in a column neighboring the specific column, an area containing the smallest number of particles is selected from those areas adjacent the area selected in the specific column. This operation for the neighboring column is thenceforth successively repeated for all subsequent columns to select the area containing the smallest number of particles in each column. Finally, a representative point is set in each area selected in each column, and an approximation curve is obtained by connecting each of the points. The approximation curve serves as the discriminant.

    摘要翻译: 通过在颗粒的二维分布的图上绘制判别式,将经过测量的样品中的一个颗粒组与样品中的另一个颗粒组分界。 为了绘制判别式,将二维分布划分为以多列布置的多个区域的网格图案,确定每个区域中包含的粒子数,并选择包含最小数量的粒子的区域 从特定列中的多个区域。 接下来,在与特定列相邻的列中,从与特定列中选择的区域相邻的区域中选择包含最小数量的颗粒的区域。 对于所有后续列,连续重复该相邻列的该操作,以选择每列中包含最小数量的粒子的区域。 最后,在每列中选择的每个区域中设置代表点,并且通过连接每个点来获得近似曲线。 近似曲线用作判别式。

    Method of demarcating one-dimensional distribution
    3.
    发明授权
    Method of demarcating one-dimensional distribution 失效
    划分一维分布的方法

    公开(公告)号:US5117357A

    公开(公告)日:1992-05-26

    申请号:US279950

    申请日:1988-12-05

    申请人: Takanobu Inoue

    发明人: Takanobu Inoue

    CPC分类号: G01N15/1456 G06F19/366

    摘要: A method for demarcating a reticulocyte particle group from a red blood cell particle group in a one-dimensional frequency distribution assumes a normal distribution on one side of the distribution. Various positions Xs.sub.1, Xs.sub.2, . . . , Xs.sub.n, which respectively indicate frequencies that are p.sub.1, p.sub.2, . . . p.sub.n percent (where n is an integer of 3 or greater) of a peak frequency of the distribution, on the normal side of the distribution, are determined. An average value and standard deviation of the normal distribution is determined from these values. A threshold value is next obtained from the average value and standard deviation, the threshold value being used to separate the reticulocyte group from the red blood cell group. The average value and standard deviation may be obtained using a least square method or by computing the mean of several average values and standard deviations.

    摘要翻译: 在一维频率分布中从红细胞粒子群划分网织红细胞粒子群的方法在分布的一侧呈现正态分布。 各种位置Xs1,Xs2,。 。 。 ,Xsn,分别表示p1,p2,...的频率。 。 。 确定在分布的正常侧上的分布的峰值频率的pn百分比(其中n是3或更大的整数)。 从这些值确定正态分布的平均值和标准偏差。 接下来从平均值和标准偏差获得阈值,阈值用于将网织红细胞组与红细胞组分离。 可以使用最小二乘法或通过计算几个平均值和标准偏差的平均值来获得平均值和标准偏差。