JPH0472547A - Method for judging cohesive image - Google Patents

Abstract

PURPOSE:To obtain accurate results of judgement with high reliability by extracting data on the measurement of a boundary portion from data on the measurement of reaction patterns so as to find the rate of change of the data on the measurement of the boundary portion, and automatically judging whether or not particles to be detected cohere, or judging other attributes of the particles according to the rate of change. CONSTITUTION:An image processing circuit 16 converts data on the image of the bottom face of a well 11a input from a video camera 15 into digital data. Data on the area of a preset well center portion 17 is taken out by a fixed amount against an image converted into the digital data and the average value C of the data is found and also the average value P of data on the area of a well peripheral portion 18 is found. Thereafter, a preset positive value is added to the average value C so as to find (c) and also a preset positive value is subtracted from the average value P so as to find (p). Data satisfying the relation of p>x>c is extracted from the data on the area of the well center portion 17 so as to extract data on the boundary portion between the center portion and peripheral portion of a reaction pattern indicated by the arrow D and the data extracted is subjected to two- or one-dimensional differential processing and the rate of change of the amount of light transmitted through the boundary portion is found and its average value X is found. The average value X is fed to a data processing circuit 19 and is compared to a preset reference value so as to judge whether or not particles cohere.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention is used in clinical tests, etc. to optically measure the reaction pattern of test particles formed on the bottom of a reaction container to determine if they are agglomerated, non-agglomerated, or The present invention relates to an agglomerated image determination method for automatically determining other attributes.

[Conventional technology]

As a conventional agglomerated image determination method, for example, Japanese Patent Application Laid-open No. 58-10
As disclosed in Publication No. 5065, the ratio of brightness between the center and the periphery is determined from the measurement data of the reaction pattern of the particles to be tested, and aggregation or non-aggregation is determined based on that ratio. Also, as disclosed in Japanese Patent Laid-Open Nos. 61-215948, 62-105031, 63-58237, and 63-256839, the optical state of the TV left camera microplate is captured. The image data is processed to determine the area of particles that are agglomerated or that have fallen to low areas such as the center of the reaction vessel using the static method, and based on that area, it is determined whether the particles are agglomerated or not. There is something like this.

[Problem to be solved by the invention]

However, in the conventional aggregation image determination method described above, the agglomeration force is very weak, and when a reaction pattern that is almost the same as a non-aggregation pattern is formed, it is very difficult to automatically and accurately determine this. becomes difficult. As described above, in the conventional agglutinated image determination method described above, the reliability of the determination result is low, so in the automatic reaction pattern determination device that implements such agglutinated image determination method, it is necessary to manipulate the determination result. There is a problem in that the operator or the like needs to check and correct each correction by visual inspection or other means, which increases the burden on the operator and the like.

This invention was made by focusing on these conventional problems, and it is possible to automatically and accurately determine even agglomeration patterns with very weak cohesive force, and thus to obtain highly reliable determination results. The purpose is to provide a determination method.

[Means to solve the problem]

In order to achieve the above object, the present invention optically measures the reaction pattern of the test particles formed on the bottom surface of a reaction vessel, and based on the measurement data, determines whether the test particles are agglomerated, non-agglomerated, or otherwise. In automatically determining the attributes of 5. The measurement data of the boundary between the center and the periphery of the reaction pattern is extracted from the measurement data, the rate of change is determined, and based on the rate of change, it is determined whether the test particles are agglomerated, non-agglomerated, or other attributes. Automatically determine.

[Effect]

As shown in FIGS. 1A and B, a test solution containing test particles is placed in a reaction container 1 having a conically concave bottom such as a microplate, and a reaction pattern is formed by, for example, a standing method. Non-aggregation pattern shown in Figure 1A and Figure 1B
In some cases, the agglomeration pattern shown in Figure 1 shows a very weak cohesive force, or almost the same pattern. However, in the agglomeration pattern with a weak cohesive force shown in FIG. 1B, the boundary of the test particles settled in the center of the reaction vessel 1 is blurred and clearly compared to the non-aggregation pattern shown in FIG. 1A. There is a characteristic that it does not.

In this invention, such characteristics are utilized to distinguish between non-agglomerated patterns and agglomerated patterns with weak cohesive force.

That is, in FIGS. 1A and 1B, for example, when the amount of transmitted light on a straight line passing through the center of the reaction vessel 1 is measured, it becomes as shown in FIGS. 2A and 2B, respectively. Now, if we focus on the boundary between the center and the periphery of the reaction pattern shown by the arrows in Figures 2A and B, the rate of change in the amount of transmitted light in this area is greater than that of the non-agglomerated pattern in Figure 2A. It can also be seen that the agglomeration pattern with weak cohesive force shown in FIG. 2B is smaller. Therefore, by extracting the measurement data of the boundary between the center and the periphery of the reaction pattern from the measurement data of the amount of transmitted light and checking the rate of change of the measurement data at the boundary, it is possible to identify an aggregation pattern with weak cohesion. It becomes possible to accurately distinguish between the pattern and the non-agglomerated pattern.

Note that the above rate of change is calculated based not only on the measurement data of the amount of transmitted light on a straight line passing through the center of the reaction vessel l, but also on the boundary between the center and the periphery of the reaction pattern by incorporating image data of the bottom of the reaction vessel. It is also possible to extract part or all of the image data of the part and process the extracted image data. In addition, for the boundary between the center and the periphery of the reaction pattern, an intermediate brightness is calculated based on the brightness of the center and periphery, and the area with the intermediate brightness is used as the boundary. It can also be detected. In this case as well, by checking the rate of change of the intermediate brightness measurement data, it is possible to accurately distinguish between agglomerated patterns with weak cohesive force and non-agglomerated patterns.

1: Embodiment] FIG. 3 is a block diagram showing the configuration of an example of an automatic aggregation image determination apparatus that implements the present invention. In this embodiment, a microplate 11 is used as a reaction vessel, and the microplate 11 is illuminated from the bottom side by a fluorescent lamp 13 connected to a fluorescent lamp power source 12.

As shown in FIG. 4, the microplate 11 is composed of a large number of wells 11a each having a conical bottom surface formed in a matrix, each of which contains a test solution containing test particles. A reaction pattern is formed on the bottom surface by a standing method.

Images of the bottom surface of each well 11 of the microplate 11 illuminated by the fluorescent lamp 13 are sequentially captured by a video camera 15, and the image data is supplied to an image processing circuit 6.
Here, based on the input image data, the average value of the rate of change in the amount of transmitted light at the boundary between the center and the periphery of the bottom image of the well lla is determined. Note that the image data of the bottom surface of each well lla is sequentially captured by moving the microplate 11 and the video camera 15 relatively in a two-dimensional direction within a horizontal plane.

Data processing in this image processing circuit 16 will be explained below.

The image processing circuit 16 first converts input image data of the bottom surface of the well lla from the video camera 15 into digital data. Note that the conversion of this input image data into digital data is performed such that the value of bright data is large and the value of dark data is small. Next, from the image converted into digital data, a certain amount of data in the area of the well center 17 set in advance as shown in FIG. Find the average value P of the data. Thereafter, a predetermined positive value is added to the average value C to obtain C, and a predetermined positive value is subtracted from the average value P to obtain p.

Next, by extracting data that satisfies the relationship I)>X>C from the data in the area of the well center 17 in FIG. Data on the boundary between the area and the peripheral area is extracted, two-dimensional or one-dimensional differentiation processing is performed on the extracted data to determine the rate of change in the amount of transmitted light at the boundary, and the average value X thereof is determined.

The average value X of the rate of change in the amount of transmitted light at the boundary, determined by the image processing circuit 16 as described above, is supplied to the data processing circuit 19, where it is compared with a predetermined reference value to determine aggregation/non-aggregation. The judgment result is displayed on the display section 21 in response to an instruction from the input section 20 such as a keyboard.

In this way, data on the boundary between the center and periphery of the reaction pattern is extracted from the image data of the bottom surface of well lla, the rate of change is determined, and based on the rate of change, aggregation/non-aggregation of the reaction pattern is determined. If this is done, even agglomerated patterns with weak cohesion will not be mistaken for non-agglomerated patterns.
Judgment can be made automatically and accurately, and highly reliable judgment results can be obtained. Just like in the past,
Since there is no need for the operator or the like to check and correct the determination result visually or by other means, the burden on the operator or the like can be significantly reduced.

In this embodiment, the microplate 11 and the video camera 15 are moved relatively in a two-dimensional direction within a horizontal plane to sequentially capture image data of the bottom surface of each well 11a of the microplate 11. The image data of the entire microplate 11 may be taken in, and the image data of the bottom surface of each well lla' may be extracted from the image data and processed in the same way.

FIG. 6 is a block diagram showing the configuration of another example of an automatic agglomerated image determination apparatus embodying the present invention. In this embodiment, the microplate 11 is spot-illuminated from the bottom side through a lens group 33 by a light source 32 connected to a power source 31.
The transmitted light is received by a light receiver 34. The output of the light receiver 34 is converted into a digital signal by a light reception data processing section 35 and supplied to a data processing section 36 . Further, the microplate 11 is moved linearly in a horizontal plane via the microplate transfer mechanism 37 under the control of the data processing unit 36, thereby moving the well 11a to the diameter 1a as shown in FIG. By scanning in the direction, data on the amount of transmitted light as shown in FIG. 7B is obtained. Note that the analog-to-digital conversion in the received light data processing section 35 is performed so that the value of bright data becomes large and the value of dark data becomes small.

In this embodiment, data on the amount of transmitted light as shown in FIG. The average value of the rate of change in the amount of transmitted light at the boundary between the center and the periphery of the pattern is determined.

Data processing in the data processing section 36 will be explained below.

The data processing unit 36 first calculates the average value C of the data at the center of the well set in advance for the data of each well lla, and also calculates the average value P of the data at the periphery of the well. Thereafter, the average value C is multiplied by a predetermined positive value greater than 1 to obtain C, and the average value P is multiplied by a predetermined positive value smaller than 1 to obtain p. however,
The values by which the average values C and P are respectively multiplied are values that satisfy the condition p>c. Next, by extracting data that satisfies p>x>c from the data of the well lla, data of the boundary between the center and the periphery of the reaction pattern shown by arrows in FIG. 2A and B is extracted. is extracted, differential processing is performed on the extracted data, the rate of change in the amount of transmitted light at the boundary is determined, and the average value X is determined.

Thereafter, the average value X is compared with a predetermined reference value to determine aggregation/non-aggregation.

After determining the reaction pattern in the data processing section 36 as described above, the determination result is transferred to the input section 3 such as a keyboard.
8 is displayed on the display section 39 in response to an instruction from 8.

In this way, the data on the boundary between the center and the periphery of the reaction pattern is extracted from the data representing the transmitted light amount distribution in the diameter direction of well lla, the rate of change is determined, and the reaction pattern is calculated based on the rate of change. If agglomeration/non-aggregation is determined, even an agglomeration pattern with weak cohesive force can be automatically and accurately determined without being mistaken as a non-aggregation pattern, as in the above-mentioned embodiment, and reliability can be improved. High judgment results can be obtained. Therefore, it is not necessary for the operator or the like to check and correct the determination result visually or by other means, as in the past, and the burden on the operator or the like can be significantly reduced.

〔Effect of the invention〕

As described above, according to the present invention, the boundary between the center and the periphery of the reaction pattern is blurred and unclear in the agglomerated pattern, which has a weaker cohesive force than the non-agglomerated pattern. Focusing on the measurement data of the reaction pattern, extract the measurement data of the boundary part mentioned above, find the rate of change, and automatically determine whether the test particles are agglomerated, non-agglomerated, or other attributes based on the rate of change. As a result, highly reliable and accurate determination results can be obtained.

Therefore, in the apparatus embodying the present invention, there is no need for the operator or the like to check and correct the determination result visually or by other means, as is the case in the past, so that the burden on the operator or the like can be significantly reduced.

[Brief explanation of the drawing]

1A, B and 2A, B are diagrams for explaining the present invention in detail; FIG. 3 is a block diagram showing the configuration of an example of an automatic agglutination image determination device implementing the present invention; This figure shows the structure of the microplate shown in FIG. 3, FIG. 5 is a diagram for explaining the operation of FIG. 3, and FIG. A block diagram showing the configuration and FIGS. 7A and 7B are diagrams for explaining its operation. 1... Reaction container 11... Microplate 11a... Well 12... Fluorescent light power source 13... Fluorescent light 15... Video camera 16... Image processing circuit 17... Well center 18...Well peripheral area 19...Data processing circuit 20...Input section 21...Display section 31
... Power source 32... Light source 33... Lens group 34... Light receiver 35... Light reception data processing section 36... Data processing section 37... Microplate transfer mechanism

Claims (1)

[Claims] 1. In optically measuring the reaction pattern of the test particles formed on the bottom of the reaction vessel and automatically determining whether the test particles are agglomerated, non-agglomerated, or other attributes based on the measurement data. , Extract the measurement data of the boundary between the center and the periphery of the reaction pattern from the measurement data, determine the rate of change, and determine whether the test particles are agglomerated, non-agglomerated, or otherwise based on the rate of change. A method for determining agglomerated images characterized by automatically determining attributes.