JPS60250255A - White corpuscle sorter - Google Patents
White corpuscle sorterInfo
- Publication number
- JPS60250255A JPS60250255A JP59106378A JP10637884A JPS60250255A JP S60250255 A JPS60250255 A JP S60250255A JP 59106378 A JP59106378 A JP 59106378A JP 10637884 A JP10637884 A JP 10637884A JP S60250255 A JPS60250255 A JP S60250255A
- Authority
- JP
- Japan
- Prior art keywords
- image
- output
- white blood
- blood cell
- concentration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000366 juvenile effect Effects 0.000 claims abstract description 3
- 210000000265 leukocyte Anatomy 0.000 claims description 14
- 210000003979 eosinophil Anatomy 0.000 claims description 9
- 210000000440 neutrophil Anatomy 0.000 claims description 7
- 210000000601 blood cell Anatomy 0.000 claims description 5
- 210000003743 erythrocyte Anatomy 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims 2
- 230000015654 memory Effects 0.000 abstract description 15
- 210000000805 cytoplasm Anatomy 0.000 abstract description 5
- 238000004458 analytical method Methods 0.000 abstract description 2
- 238000003556 assay Methods 0.000 abstract 1
- 230000003448 neutrophilic effect Effects 0.000 abstract 1
- 238000000034 method Methods 0.000 description 8
- 235000019640 taste Nutrition 0.000 description 4
- 238000010521 absorption reaction Methods 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 238000012303 cytoplasmic staining Methods 0.000 description 1
- 239000008187 granular material Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Optical investigation techniques, e.g. flow cytometry
- G01N15/1468—Optical investigation techniques, e.g. flow cytometry with spatial resolution of the texture or inner structure of the particle
Landscapes
- Chemical & Material Sciences (AREA)
- Dispersion Chemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
【発明の詳細な説明】
〔発明の利用分野〕
本発明は、白血球の分類に有効な特徴量に関す（１）
るもので、特に好酸球と好中球・幼若域の３者の分類識
別に好適な白血球分類装置に関する。[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to (1) feature values that are effective for classifying white blood cells, and in particular, the characteristics of the three types of white blood cells, eosinophils, neutrophils, and immature cells. The present invention relates to a leukocyte classification device suitable for classification and identification.
血球像の入力に関する色分解の波長としては、Ｒ： ５
６０〜６００ｎｍ、Ｇ： ５１０〜５４０ｎｍ、Ｂ：
４００〜４３０ｎｍが適している。従来Ｇとして５５０
ｎｍのフィルタ（以下、ｇ′と記す）を用いていた（特
願昭５０−１０５９０６号明細書参照）。このフィルタ
で入力した白血球像（入力した血球像からＢのフィルタ
を介して入力した血球像をもとに赤血球像を除いたもの
）の濃度分布は、今回選定した３フィルタｒ：５９０ｎ
ｍ＊ｇ：５２５ｎｍ、ｂ ： ４１５ｎｍのうちｒで入
力した白血球像と同様の濃度分布（第１図（ａ））を示
す。The color separation wavelength for blood cell image input is R: 5.
60-600nm, G: 510-540nm, B:
400-430 nm is suitable. 550 as conventional G
A nm filter (hereinafter referred to as g') was used (see the specification of Japanese Patent Application No. 50-105906). The concentration distribution of the white blood cell image input with this filter (the red blood cell image is removed based on the input blood cell image input through filter B) is as follows: 3 filters selected this time: 590n
m*g: 525 nm, b: 415 nm, showing the same concentration distribution as the white blood cell image input with r (FIG. 1(a)).
従来は、この第１図（ａ）の濃度分布Ｃ以下、濃度とヒ
ストグラムと記す）から、白血球の核と細胞質を濃度で
分離する閾値４，５，６．白血球の細胞質と背景を濃度
で分離する閾値７，８．９をめ、各々の平均閾値１７，
１８，１９をめ（２）
る。この平均閾値１７，１８．１９以上の濃度をもつ総
頻度と、核と細胞質間の閾値４，５．６以上の濃度をも
つ総頻度（領域１，２．３・・・白血球の核面積に相当
する）の比から好中球、幼若味。Conventionally, thresholds 4, 5, 6, . The threshold values for separating white blood cell cytoplasm and background by concentration are 7 and 8.9, and the average threshold values for each are 17 and 8.9, respectively.
18 and 19 (2). The total frequency with a concentration above the average threshold value of 17, 18.19, and the total frequency with a concentration above the threshold value of 4, 5.6 between the nucleus and the cytoplasm (area 1, 2.3... The nuclear area of white blood cells Neutrophils, immature taste from the ratio of corresponding).
好酸球の３者を分類していた。しかし必ずしも幼若味と
好酸球の分類には有効でないという欠点があった。Three types of eosinophils were classified. However, it has the drawback that it is not necessarily effective in classifying juvenile taste and eosinophils.
本発明の目的は、上記欠点を解消し、好中球。 The object of the present invention is to eliminate the above-mentioned drawbacks and to improve the production of neutrophils.
幼若球、好酸球の３者をより精度良く識別する白血球分
類装置を提供することにある。It is an object of the present invention to provide a white blood cell classification device that can more accurately distinguish between immature cells and eosinophils.
従来の手法は、単波長における濃度ヒストグラムの分布
の相異から、特徴パラメータを定量化し分類を行なうも
のであった。これに対し、今回は細胞質（細胞質内に含
まれる顆粒も含む）の染色性を最も強調するフィルタ（
ｒ：５９０ｎｍ、ｇ： ５２５ ｎ ｍ ｔ ｂ ：
４１５ ｎ ｍ）を選択した利点を活かし、単波長にお
ける濃度ヒストグラムでなく、複数波長間の濃度ヒスト
グラムから定量化で（３）
きる特徴パラメータをもとに、より精度良く分類を行お
うというものである。Conventional methods quantify characteristic parameters and perform classification based on differences in the distribution of concentration histograms at single wavelengths. In contrast, this time we used a filter (
r: 590 nm, g: 525 nm tb:
Taking advantage of the advantage of selecting 415 nm), we aim to perform more accurate classification based on characteristic parameters that can be quantified (3) from concentration histograms between multiple wavelengths, rather than concentration histograms at a single wavelength. be.
第１図に、ｒおよびｇを介して入力した白血球像の濃度
ヒストグラムを示している。FIG. 1 shows a density histogram of a white blood cell image input via r and g.
ここでは、好酸球の顆粒（細胞質上に密に存在する）の
ｇ：５２５ｎｍにおける吸収率が、核の吸収率にほぼ等
しくなることを利用している。すなわち、ｇの濃度ヒス
トグラムから白血球の核とみなされる部分１３に対する
、ｒの濃度ヒストグラムから核とみなされる部分１３の
比が、好中球におけるｒの核１の、ｇの核１１に対する
比、幼若味ｒの核２の、ｇの核１２に対する比に較べて
小さく（第２図）なり、好酸球と他の２者の分類識別が
可能となる。Here, it is utilized that the absorption rate of eosinophil granules (densely present on the cytoplasm) at g:525 nm is approximately equal to the absorption rate of the nucleus. That is, the ratio of the part 13 considered to be the nucleus from the concentration histogram of r to the part 13 considered to be the nucleus of white blood cells from the concentration histogram of g is the same as the ratio of nucleus 1 of r to nucleus 11 of g in neutrophils, The ratio of nucleus 2 of young taste r to nucleus 12 of g is smaller (FIG. 2), making it possible to classify eosinophils and the other two.
以下、本発明の実施例を第３図、第４図により説明する
。Embodiments of the present invention will be described below with reference to FIGS. 3 and 4.
血球像を赤フィルタ３１、緑フィルタ４１、青フィルタ
５１を介して光電変換器３２，４２゜５２で電気信号に
変換する。ここで、光電変換器（４）
としでは例えば撮像管、固体撮像素子等が考えられる。The blood cell image is passed through a red filter 31, a green filter 41, and a blue filter 51, and then converted into an electrical signal by photoelectric converters 32, 42°52. Here, the photoelectric converter (4) may be, for example, an image pickup tube, a solid-state image sensor, or the like.
ついで上記光電変換器３２，４２，５２の出力を各々Ａ
／Ｄ変換器３３，４３，５３でディジタル化し、画像メ
モリ３４，４４，５４に記憶する。上記画像メモリ３４
，４４．５４のデータをもとに特徴抽出装置３５におい
て、濃度ヒストグラムを生成し、演算識別装置３６で濃
度ヒストグラムの解析を行い、その解析結果をもとに白
血球を分類する。Then, the outputs of the photoelectric converters 32, 42, and 52 are respectively A.
/D converters 33, 43, 53 digitize and store in image memories 34, 44, 54. The above image memory 34
, 44.54, the feature extraction device 35 generates a concentration histogram, the arithmetic identification device 36 analyzes the concentration histogram, and the white blood cells are classified based on the analysis results.
第４図は、上記特徴抽出装置３５の一構成例を示したも
のである。FIG. 4 shows an example of the configuration of the feature extracting device 35. As shown in FIG.
画像メモリ３４の出力および画像メモリ５４の出力およ
び定数値を記憶したメモリ６２の出力を引算回路６１に
導き、画像メモリ３４の出力から画像メモリ５４の出力
を引いた信号を作る。一方２値化回路６３．６４にあら
かじめめている閾値を演算識別装置３６から与えておく
。２値化回路６３には、画像メモリ５４の出力を導き、
設定閾値以上の信号の場合は′１′″、それ以外の場合
は０”の信号を出力する。また２値化回路６４には、（
５）
上記引算回路６１の出力を導き、設定閾値以下の信号の
場合はｕ １ ｐｔ、それ以外の場合はａｔ Ｏ＃ｌの
信号を出力する。ついで、上記２値化回路６３，６４の
＃＃Ｑｊｌ、”ｌ”信号をＡＮＤ回路６５に導き、血球
像中の赤血球部分は１１１７′、その他の部分は′０″
の２値化信号を作る（詳細は特願昭５７−２２３８２０
号″領域分割方法″を参照）。The output of the image memory 34, the output of the image memory 54, and the output of the memory 62 storing constant values are led to a subtraction circuit 61, and a signal obtained by subtracting the output of the image memory 54 from the output of the image memory 34 is generated. On the other hand, a predetermined threshold value is given to the binarization circuits 63 and 64 from the arithmetic identification device 36. The output of the image memory 54 is led to the binarization circuit 63,
If the signal is equal to or higher than the set threshold, a signal of ``1'' is output, otherwise a signal of 0'' is output. Furthermore, the binarization circuit 64 has (
5) The output of the subtraction circuit 61 is derived, and if the signal is below the set threshold, the signal u 1 pt is output, otherwise the signal at O#l is output. Next, the ##Qjl and "l" signals of the binarization circuits 63 and 64 are led to the AND circuit 65, and the red blood cell part in the blood cell image is set to 1117', and the other parts are set to '0'.
Create a binary signal (see patent application No. 57-223820 for details)
(Refer to No. ``Region Division Method'').
一方、選択回路６６は、演算識別装置３６の信号により
、画像メモリ３４あるいは画像メモリ４４のいずれか一
方を選択するように設定し、選択した一方の信号をゲー
ト回路６７に出力する。On the other hand, the selection circuit 66 is set to select either the image memory 34 or the image memory 44 according to the signal from the arithmetic identification device 36, and outputs the selected signal to the gate circuit 67.
ゲート回路６７のゲート信号として上記ＡＮＤ回路６５
の出力を用い、赤血球部分に対応する入力信号の通過を
禁止した信号を作る。上記ゲート回路６７の出力をヒス
トグラム生成回路６８に導けば、所望の濃度ヒストグラ
ムを得ることができる。The above AND circuit 65 serves as a gate signal for the gate circuit 67.
Using the output of , a signal is created that prohibits the passage of the input signal corresponding to the red blood cell part. By guiding the output of the gate circuit 67 to the histogram generation circuit 68, a desired density histogram can be obtained.
選択回路６６の指定を変えて、２回同じ操作を繰り返す
ことにより、ｒの濃度ヒストグラムとｇの濃度ヒストグ
ラムがまる。By changing the designation of the selection circuit 66 and repeating the same operation twice, the density histogram of r and the density histogram of g are matched.
上記、濃度ヒストグラムを演算識別装置３６に（６）
導き、核と細胞質の閾値ペア（４，１４）、（５゜１５
）、（６，１６）をめる。その手法は、濃度の高い方の
谷を検出することによりめる。上記各のない場合は濃度
の低い方の谷と濃度の高い方の山の濃度値を一定の比率
で内分した濃度値を閾値とする（詳細は特願昭５７−９
４３８１号″閾値決定法″′を参照）。ついで、上記閾
値ペア（４，１４）。The concentration histogram described above is guided to the calculation discriminator 36 (6), and the nucleus and cytoplasm threshold pairs (4, 14), (5°15
), (6,16). The method is determined by detecting the valley with higher concentration. If none of the above exists, the threshold value is the concentration value obtained by internally dividing the concentration value of the valley with lower concentration and the concentration value of the mountain with higher concentration at a certain ratio (for details, see Japanese Patent Application No. 57-9
No. 4381 ``Threshold Determination Method''). Next, the threshold pair (4, 14).
（５，１５）、（６，１６）から、核面積相当値のペア
（１，１１）、（２，１２）、（３，ｔ３）をめ、各ペ
アの比から入力された白血球の分類を行う。ここでは、
演算識別装置３６として汎用の計算機を想定してその処
理手順を述べたが、この処理をハード化し高速化するこ
とも可能である。From (5, 15) and (6, 16), find the pair of nuclear area equivalent values (1, 11), (2, 12), (3, t3), and classify the white blood cells input from the ratio of each pair. I do. here,
Although the processing procedure has been described assuming a general-purpose computer as the arithmetic identification device 36, it is also possible to implement this processing in hardware and speed it up.
また、全体の構成としても、Ａ／Ｄ変換器を１つだけも
って光電変換器の出力を逐次切替えてＡ／Ｄ変換する方
式、光電変換器・Ａ／Ｄ変換器を１系統だけもって、フ
ィルタを切替える方式、画像メモリを用いず、必要な画
像をその都度Ａ／Ｄ変換する方式等変形は多々可能であ
る。In addition, the overall configuration is a method that has only one A/D converter and sequentially switches the output of the photoelectric converter for A/D conversion, and a method that has only one system of photoelectric converters and A/D converters and filters. Many modifications are possible, such as a method of switching between images, a method of A/D converting a necessary image each time without using an image memory, etc.
（７）
本発明によれば、好中球、幼若味、好酸球の細胞質の染
色性の違いを強調した特徴パラメータの定量化ができる
ので、上記３者の分類精度向上に多大の効果を得ること
ができる。(7) According to the present invention, it is possible to quantify characteristic parameters that emphasize the differences in cytoplasmic staining of neutrophils, immature tastes, and eosinophils, which has a great effect on improving the classification accuracy of the above three types. can be obtained.
第１図ａ、ｂは、好中球、幼若味、好酸球の赤および緑
光下での濃度ヒストグラム、第２図は特徴パラメータの
分布範囲を示す図、第３図、第４図は本発明の一実施例
になる装置の構成を示すブロック図である。
３１．４１，５１・・・光学フィルタ、３２，４２゜５
２・・・光電変換装置、３３，４３，５３・・・Ａ／Ｄ
変換器、３４，４４，５４・・・画像メモリ、３５・・
・特徴抽出装置、３６・・・演算識別装置、６１・・・
引算回路、６２・・・メモリ、６３，６４・・・２値化
回路、６５・・・ＡＮＤ回路、６６・・・選択回路、６
７・・・ゲー（８）
子 ２ 図
１ ／
第 ３ 目
第４− 図Figures 1a and b are concentration histograms of neutrophils, juveniles, and eosinophils under red and green light, Figure 2 is a diagram showing the distribution range of characteristic parameters, Figures 3 and 4 are FIG. 1 is a block diagram showing the configuration of an apparatus according to an embodiment of the present invention. 31.41,51...Optical filter, 32,42゜5
2... Photoelectric conversion device, 33, 43, 53... A/D
Converter, 34, 44, 54... Image memory, 35...
-Feature extraction device, 36... Arithmetic identification device, 61...
Subtraction circuit, 62... Memory, 63, 64... Binarization circuit, 65... AND circuit, 66... Selection circuit, 6
7... Game (8) Child 2 Figure 1 / 3rd eye 4th - Figure
Claims (1)
ｂ画像と略す）を用いて、５画像と、１画像と５画像の
差画像から赤血球を検出する検出手段と、ｇ画像あるい
は１画像の一方を選択する選択手段を有し、上記選択手
段の出力を上記検出手段の出力で信号の通過を制御する
制御手段を有し、上記制御手段の出力をもとにヒストグ
ラムを生成する生成手段を有し、上記生成手段の出力を
もとに１画像あるいはｇ画像での白血球の核とみなされ
る値を算出する演算手段と、上記演算手段の２つの出力
の比から、白血球特に好酸球と好中球。 幼若域を識別する識別手段を有する白血球分類装置。[Claims] Red components input by wavelength by photoelectric conversion means. Blood cell images of green component and blue component (hereinafter, 1 image 2g image each,
a detection means for detecting red blood cells from 5 images and a difference image between the 1 image and the 5 images; and a selection means for selecting either the g image or the 1 image; The output includes a control means for controlling the passage of a signal using the output of the detection means, and a generation means for generating a histogram based on the output of the control means. Alternatively, white blood cells, particularly eosinophils and neutrophils, can be determined from the calculation means for calculating the value considered to be the nucleus of white blood cells in the g image and the ratio of the two outputs of the above calculation means. A white blood cell classification device having an identification means for identifying a juvenile region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP59106378A JPS60250255A (en) | 1984-05-28 | 1984-05-28 | White corpuscle sorter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP59106378A JPS60250255A (en) | 1984-05-28 | 1984-05-28 | White corpuscle sorter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPS60250255A true JPS60250255A (en) | 1985-12-10 |
Family
ID=14432049
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP59106378A Pending JPS60250255A (en) | 1984-05-28 | 1984-05-28 | White corpuscle sorter |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS60250255A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS63290946A (en) * | 1987-05-22 | 1988-11-28 | Shimadzu Corp | Cell identification device |
-
1984
- 1984-05-28 JP JP59106378A patent/JPS60250255A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS63290946A (en) * | 1987-05-22 | 1988-11-28 | Shimadzu Corp | Cell identification device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US3999047A (en) | Method and apparatus utilizing color algebra for analyzing scene regions | |
| JP6959614B2 (en) | Analyzer and flow cytometer | |
| US3851156A (en) | Analysis method and apparatus utilizing color algebra and image processing techniques | |
| JP7428994B2 (en) | measurement system | |
| KR20190043135A (en) | Systems and methods for classifying biological particles | |
| US20170102310A1 (en) | Flow cytometer and a multi-dimensional data classification method and an apparatus thereof | |
| US20080172185A1 (en) | Automatic classifying method, device and system for flow cytometry | |
| JPH0658928A (en) | Particle analyzer | |
| KR101995763B1 (en) | Cytometry apparatus and method | |
| CN107389536B (en) | Flow cell particle classification counting method based on density-distance center algorithm | |
| Frankel et al. | Application of neural networks to flow cytometry data analysis and real‐time cell classification | |
| CN106975617B (en) | A kind of Classification of materials method based on color selector | |
| CN109446964A (en) | Face detection analysis method and device based on end-to-end single-stage multiple scale detecting device | |
| KR101995764B1 (en) | Cytometry apparatus and method | |
| CN114332856A (en) | Feature extraction method and device for blood type card image and blood type interpretation system | |
| US4850024A (en) | Method and apparatus for classifying white blood cells | |
| JP4554810B2 (en) | Particle analyzer and particle analysis method | |
| US5684584A (en) | Apparatus for analyzing cells in urine | |
| JP3002325B2 (en) | Surface inspection equipment | |
| JPS60250255A (en) | White corpuscle sorter | |
| JP3504030B2 (en) | Method and apparatus for determining particle criterion, and particle analyzer using the criterion | |
| JPH02194359A (en) | Separation of blood cell of three main types from leucocyte cell histogram | |
| JPS5830049B2 (en) | Automatic reticulocyte measurement device | |
| JPH0443311B2 (en) | ||
| Selvaraj et al. | ‘K-Means Clustering Based Segmentation of Lymphocytic Nuclei for Acute Lymphocytic Leukemia Detection |