公开(公告)号：US5162204A

公开(公告)日：1992-11-10

申请号：US347219

申请日：1989-05-04

申请人： Harumi Matsuzaki ,  Ryoichi Haga ,  Yuusaku Nishimura ,  Kenji Baba ,  Fumio Maruhashi ,  Nobuko Nishimura ,  Masahiko Ishida ,  Setuo Saitoh

发明人： Harumi Matsuzaki ,  Ryoichi Haga ,  Yuusaku Nishimura ,  Kenji Baba ,  Fumio Maruhashi ,  Nobuko Nishimura ,  Masahiko Ishida ,  Setuo Saitoh

IPC分类号： C12M1/34 ,  C12M3/00 ,  C12M3/02 ,  C12M3/06 ,  C12N1/12 ,  C12N1/20 ,  C12N5/02 ,  G01N33/50

CPC分类号： G01N33/5005 ,  C12M41/36 ,  C12M41/46 ,  C12N1/12 ,  C12N1/20 ,  Y10S435/808 ,  Y10S435/813

摘要： An apparatus for culturing animal cells, includes a diagnosis system to make it possible to diagnose by measuring sizes of the cells by means of image processing while a system of culturing the animal cells is cut off from the outside of the apparatus system, a control system to control the culture conditions of a culturing cistern on the basis of a result of the diagnosis, and a culture medium regenerating system to remove waste contents in a culture medium. In the apparatus of this invention, the culture system and diagnosis system are cut off from the outside of the apparatus system. Therefore, no infection of miscellaneous microorganism, etc., occurs, nor does consumption of valuable cells and products occur. Further, since it is possible to easily obtain data of the proportions of division potential-possessing cells and living cells, the environmental factors to the culturing cistern can be easily and instantly controlled based on the data, and therefore, the culturing efficiency can be enhanced.

公开(公告)号：US5403735A

公开(公告)日：1995-04-04

申请号：US735625

申请日：1991-07-24

申请人： Fumio Maruhashi ,  Nobuko Nishimura ,  Ryoichi Haga ,  Harumi Matsuzaki ,  Ryusei Nakano

发明人： Fumio Maruhashi ,  Nobuko Nishimura ,  Ryoichi Haga ,  Harumi Matsuzaki ,  Ryusei Nakano

IPC分类号： C12M1/36 ,  C12M1/34 ,  C12N5/06 ,  G01N15/14 ,  G06T1/00 ,  G06T7/20 ,  G06T7/60 ,  C12N5/00 ,  F23N5/08

CPC分类号： C12M41/46 ,  C12M41/36 ,  G01N15/1468 ,  Y10S435/808

摘要： In order to investigate an object, such as a culture of micro-organisms, the object is repeatedly captured at two different magnifications by a suitable image pick-up device. The images at the two different magnifications are then analysed by a suitable picture image recognition device and the results of the repeated analysis are compared, thereby to derive a measurement of the change in the object with time. Thus, for a culture of micro-organisms, the number of cells can be determined at one magnification and the number of microscopic spherical bodies can be determined at another magnification, to obtain a measure of the activity of the culture. Preferably, one or more conditions of the object are then controlled on the basis of the measurement of change.

摘要翻译： 为了研究诸如微生物培养物的对象，通过合适的图像拾取装置以两种不同的放大率重复地捕获物体。 然后通过合适的图像图像识别装置分析两个不同放大倍率的图像，并比较重复分析的结果，由此导出对象随时间变化的测量。 因此，对于微生物的培养物，可以在一个放大倍数下确定细胞数，并且可以在另一放大倍数下确定微球体的数量，以获得培养物的活性的量度。 优选地，基于变化的测量来控制物体的一个或多个条件。

公开(公告)号：US5943662A

公开(公告)日：1999-08-24

申请号：US220546

申请日：1994-03-31

申请人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

发明人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

IPC分类号： G05B13/02 ,  G06F9/44 ,  G06F15/18 ,  G06N3/08 ,  G06N5/04 ,  G06N7/02

CPC分类号： G05B13/027 ,  G05B13/0285 ,  Y10S706/903 ,  Y10S706/906

摘要： A method and a system for causing a neural circuit model to learn typical past control results of a process and using the neural circuit model for supporting an operation of the process. The neural circuit model is caused to learn by using, as input signals, a typical pattern of values of input variables at different points in time and, as a teacher signal, its corresponding values of the control variable. An unlearned pattern of input variables is inputted to the thus-learned neuron circuit model, whereby a corresponding value of the control variable is determined. Preferably, plural patterns at given time intervals can be simultaneously used as patterns to be learned.

摘要翻译： 一种用于使神经电路模型学习过程的典型过去控制结果并使用所述神经电路模型来支持所述过程的操作的方法和系统。 通过使用神经电路模型作为输入信号，在不同的时间点使用输入变量的典型值，并且作为教师信号，控制变量的相应值。 将输入变量的未校正模式输入到如此学习的神经元电路模型中，由此确定控制变量的对应值。 优选地，以给定时间间隔的多个图案可以同时用作要学习的图案。

公开(公告)号：US5845052A

公开(公告)日：1998-12-01

申请号：US582018

申请日：1996-01-02

申请人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

发明人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

IPC分类号： G05B13/02 ,  G06F9/44 ,  G06F15/18 ,  G06N3/08 ,  G06N5/04 ,  G06N7/02 ,  G06K9/00

CPC分类号： G05B13/027 ,  G05B13/0285 ,  Y10S706/903 ,  Y10S706/906

摘要： A method for causing a neural circuit model to learn typical past control results of a process and using the neural circuit model for supporting an operation of the process. The neural circuit model is caused to learn by using, as input signals, a typical pattern of values of input variables at different points in time and, as a teacher signal, its corresponding values of the control variable. An unlearned pattern of input variables is inputted to the thus-learned neuron circuit model, whereby a corresponding value of the control variable is determined. Preferably, plural patterns at given time intervals can be simultaneously used as patterns to be learned.

摘要翻译： 一种用于使神经电路模型学习过程的典型过去控制结果并使用所述神经电路模型来支持所述过程的操作的方法。 通过使用神经电路模型作为输入信号，在不同的时间点使用输入变量的典型值，并且作为教师信号，控制变量的相应值。 将输入变量的未校正模式输入到如此学习的神经元电路模型中，由此确定控制变量的对应值。 优选地，以给定时间间隔的多个图案可以同时用作要学习的图案。

公开(公告)号：US5774633A

公开(公告)日：1998-06-30

申请号：US613718

申请日：1991-12-23

申请人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

发明人： Kenji Baba ,  Ichiro Enbutsu ,  Shoji Watanabe ,  Hayao Yahagi ,  Fumio Maruhashi ,  Harumi Matsuzaki ,  Hiroshi Matsumoto ,  Shunsuke Nogita ,  Mikio Yoda ,  Naoki Hara

IPC分类号： G05B13/02 ,  G06F9/44 ,  G06F15/18 ,  G06N3/08 ,  G06N5/04 ,  G06N7/02

CPC分类号： G05B13/027 ,  G05B13/0285 ,  Y10S706/903 ,  Y10S706/906

摘要： A method and a system for causing a neural circuit model to learn typical past control results of a process and using the neural circuit model for supporting an operation of the process. The neural circuit model is caused to learn by using, as input signals, a typical pattern of values of input variables at different points in time and, as a teacher signal, its corresponding values of the control variable. An unlearned pattern of input variables is inputted to the thus-learned neuron circuit model, whereby a corresponding value of the control variable is determined. Preferably, plural patterns at given time intervals can be simultaneously used as patterns to be learned.

摘要翻译： PCT No.PCT / JP90 / 00322 Sec。 371 1991年12月23日第 102（e）1991年12月23日PCT PCT 1990年3月13日PCT公布。 出版物WO90 / 10898 1990年9月20日1990年一种使神经电路模型学习过程的典型过去控制结果并使用神经电路模型支持过程操作的方法和系统。 通过使用神经电路模型作为输入信号，在不同的时间点使用输入变量的典型值，并且作为教师信号，控制变量的相应值。 将输入变量的未校正模式输入到如此学习的神经元电路模型中，由此确定控制变量的对应值。 优选地，以给定时间间隔的多个图案可以同时用作要学习的图案。