JPH06138119A - Automatic cell classifying and analyzing apparatus - Google Patents

Abstract

PURPOSE:To provide an automatic cell classifying and analyzing apparatus, which can perform stable focusing, at a low cost. CONSTITUTION:A sample, wherein a slide glass is coated with myelocytes or peripheral blood cells is magnified with an automatic microscope 3. The image is picked up with a color TV camera 6. Thus, the cells are automatically classified and analyzed in this apparatus. The image signal from the TV camera 6 is converted into the digital signal with an A/D converter 7. Each density histogram of the digital image, which is obtained every time the focal point is changed, is formed with a characteristic extracting processor 9. The difference between the most-frequent gradation value and the darkest-part gradation value for every density histogram is computed with a recognition processor 10. The maximum value among the difference values is extracted, and the corresponding focal-point position is determined as the focasing position.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is an apparatus for automatically classifying and analyzing cells by enlarging a specimen of bone marrow cells or peripheral blood cells smeared on a slide glass with a microscope and imaging the image with a color TV camera. In particular, the present invention relates to an automatic cell sorting / analyzing device characterized by automatic focusing of reticulocytes having small granules inside cells, red blood cells having micronuclei and the like.

[0002]

2. Description of the Related Art Generally, in an automatic cell sorting / analyzing apparatus, it is necessary to focus blood cells in order to obtain a clear image. As shown in FIG. 4, the in-focus position of blood cells exists on the upper surface, lower surface and granule surface of red blood cells, and automatic classification / reclassification is performed for reticulocytes having small granules inside cells, red blood cells having micronuclei, etc. For analysis, it is particularly desired to focus on the granule surface. Conventionally, when focusing on the granule surface, the more the focus is adjusted based on the waveform of the image sensor for focusing, the more the waveform becomes uneven, so the optimum position for focusing is determined by the degree of the unevenness of the waveform. A focus position is fixedly set at an intermediate position between the upper surface and the lower surface of the blood cell. Etc. are adopted.

[0003]

Among the focusing methods of the above-mentioned conventional automatic cell sorting / analyzing apparatus, since the method of using the analog signal of the entire image, the information of a small granular portion is hidden, resulting in the matching. There is a problem that the focal position is not stable, and the method of (3) relatively easily focuses on the granules, but as shown in blood cells A, B, and C of FIG. 4, the granules are always in a constant position inside the blood cells. However, for example, defocusing occurs in blood cells A and C in FIG. 4, and this method also has a problem that defocusing is likely to occur.

The present invention has been made in view of the above problems, and an object thereof is to provide an automatic cell sorting / analyzing apparatus which is inexpensive and can achieve stable focusing.

[0005]

[Means and Actions for Solving the Problems] The automatic cell sorting / analyzing device of the present invention is a microscope in which a sample prepared by smearing bone marrow cells or peripheral blood cells on a slide glass is magnified and the image is taken by a television camera. In an apparatus for automatically classifying and analyzing cells, an A / D converter that converts a video signal from the television camera into a digital signal, and a density that creates each density histogram of a digital image obtained each time the focus is changed Histogram creating means, difference value calculating means for calculating the difference between the most frequent gradation value and the darkest part gradation value for each density histogram, the maximum difference value is extracted, and the corresponding focus position is focused. A focusing position determining means for determining the position is characteristically provided.

In the automatic cell sorting / analyzing apparatus of the present invention,
To perform focusing processing, images magnified by a microscope at a plurality of focal positions are captured by a television camera, converted into digital signals by an A / D converter, and a density histogram is displayed, for example, as shown in FIG. The difference between the most frequent gradation value and the darkest part gradation value of the density histogram is calculated,
Furthermore, extract the largest of the differences in each focus position,
The focus position is determined as the focus position.

[0007]

The present invention will be described in more detail with reference to the following examples. FIG. 1 is a block diagram of an automatic cell classification / analysis apparatus in which the present invention is implemented. This automatic cell classification
In the analyzer, the automatic microscope 3 in which the sample of the sample cassette 1 is loaded / unloaded by the auto feeder 2, the focusing image sensor 5 for reading the image of the automatic microscope 3 via the half mirror 4a, and the automatic microscope 3 A color TV camera 6 for capturing an image, an A / D converter 7 for converting the captured image into a digital signal, an image memory 8 for storing digital image data, and a feature extraction processor for creating a density histogram of the digital image. 9, a recognition processor 10 for obtaining the difference between the most frequent gradation value and the darkest part gradation value of the density histogram, and a CPU 11 that executes various control operations.
, Auto feeder controller 12, and automatic microscope 3
Drive circuit 13 for moving the stage in the left-right direction back and forth, a focus drive circuit 14 for moving the lens system of the automatic microscope 3 up and down, and an automatic signal processing for focusing by a signal from the image sensor 5. Focus circuit 15, RGB monitor 16, keyboard 1
7 and a printer 18. Also the classification obtained
The analysis data is sent to the host computer 19.

Next, the focusing process of this automatic cell sorting / analyzing device will be described with reference to the flow chart shown in FIG. First, the target screen is set by operating the automatic microscope 3 with the sample attached (step ST1). Next, the image sensor 5
Then, the autofocus circuit 15 automatically focuses the focus position of the blood cells on the upper surface of the red blood cells (step ST2). The automatic focusing by the image sensor 5 is performed conventionally.

Next, an image of the sample magnified by the automatic microscope 3 is picked up by the color TV camera 6, the video signal thereof is taken in (step ST3), digitized by the A / D converter 7, and stored in the image memory 8. Yes (step ST4). In the feature extraction processor 9, the background and other
Imaged red (R), green (G), and blue (B) images (Fig. 3
Create a density histogram (see step ST)
5). An example of the density histogram for each of the three colors in this case is shown in FIG. In the figure, the right side shows the bright part and the left side shows the dark part.Since the micronucleus usually has low brightness, the more the camera catches the red blood cell micronucleus and the better the focus, the clearer the distribution in the dark part of the concentration histogram becomes. It becomes big. On the other hand, even if the micronucleus is caught, if the focus is out of focus, it will not be distributed in the darkest part, and will be a small distribution that is gentler than the most frequent gradation. Based on this, the recognition processor 10 causes the difference L between the most frequent gradation value and the darkest part gradation value of the density histogram to be L.
Is calculated (step ST6). Then, this difference value L is stored (step ST7), and it is determined whether or not the set number of times has ended (step ST8). That is, it is determined whether or not the processes of steps ST3 to ST7 have been performed for the number of times n, in which the focus is shifted from the upper surface of the red blood cells to the lower surface at a predetermined pitch for focusing. If the set number of times has not been completed yet, the determination in step ST8 is NO, and the focal position is shifted by a predetermined distance from the upper surface of the red blood cells to the lower surface (step S
T9), and returns to step ST3, the then performs steps ST3～ST7, calculates the difference value L _i of the most frequently used tone and darkest tone at the focal position. By repeating the above process n times, the difference values L ₁ , ..., L _i for n times are repeated.
..., L _n is obtained. As a result, the determination in step ST8 becomes YES, and then the maximum value is extracted from the difference values L ₁ , ..., L _n (step ST10). The position where this maximum value is obtained is determined as the focus position (step ST1).
1). In the above embodiment, the focus is moved from top to bottom, but depending on the object, from bottom to top or top,
You may make it move below.

In the above embodiment, the concentration histograms are R, G, and B, but depending on the type of the target granules, if the feature can be most emphasized, for example, red blood cells with micronuclei, It can be changed as appropriate, for example, by using a density histogram of B-G (blue and green difference image).

[0011]

According to the present invention, it is possible to always stably focus on the granule surface, so that the accuracy of the automatic classification / analysis apparatus provided by the present invention can be maintained. Further, the focus determination can be made based on the difference between the most frequent gradation value and the darkest gradation value of the histogram. Further, since all the density histograms are instantly created by hardware, the processing time is short and the cost can be low because it can be used as the hardware for automatic classification / analysis.

[Brief description of drawings]

FIG. 1 is a block diagram of an automatic cell classification / analysis device in which the present invention is implemented.

FIG. 2 is a flowchart for explaining focusing processing of the automatic cell sorting / analyzing apparatus.

FIG. 3 is a diagram showing an example of a color image of the automatic cell sorting / analyzing apparatus.

FIG. 4 is a diagram for explaining a focus position on a blood cell.

FIG. 5 is a diagram showing R, G, and B density histograms.

[Explanation of symbols]

 3 Automatic microscope 5 Image sensor 6 Color TV camera 7 A / D converter 8 Image memory 9 Feature extraction processor 10 Recognition processor

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoshihiko Kato 17 Chudoji-Minami-cho, Shimogyo-ku, Kyoto Science Center Building Co., Ltd. inside the Omron Life Science Laboratory (72) Inventor Shunji Utsunomiya 17 Minami-cho, Shimogyo-ku, Kyoto Cayenne Center Center Omron Life Science Research Institute

Claims (1)

[Claims] 1. An apparatus for automatically classifying and analyzing cells by enlarging a sample of bone marrow cells or peripheral blood cells smeared on a slide glass with a microscope and imaging the image with a TV camera, wherein the image from the TV camera is used. An A / D converter for converting the signal into a digital signal, a density histogram creating means for creating each density histogram of the digital image obtained each time the focus is changed, and a most frequent gradation value for each density histogram. A difference value calculating means for calculating a difference between the darkest tone values, and a focus position determining means for extracting a maximum difference value and determining a corresponding focus position as a focus position. Automatic cell sorting / analyzing device.