JPS58161064A - Image processing and playback device - Google Patents

Abstract

PURPOSE:To improve the reliability of the device, by making it possible to reexamine specific cells and indistinct corpuscles after the classification of all samples is terminated, and reducing the storage capacity of digital pictures of specific cells and indistinct corpuscles, and increasing the number of samples which can be reexamined. CONSTITUTION:The two-dimensional optical signal of a blood image on an optical microscope 1 is converted to an analog signal by an image sensor 3 while moving the blood image by a stage driving circuit 6 and is digitized by an A/D converter 4. Features of this digital signal are extracted by a feature extracting circuit 11 and are stored in a storage circuit 9, and a synchronizing signal from the circuit 11 is applied to a TV camera 2 and a D/A converter 10, and the signal of the circuit 9 is taken into the circuit 11 to obtain a feature quantity required for classification. When an indistinct corpuscle or an abnormal corpuscle is detected as the result of classification, the signal of the circuit 9 is stored in an external storage device 8 together with the examination object number and discrimination information by a microcomputer 12, and simultaneously, the next white corpuscle detection command is given to a digital controlling circuit 7. Thus, specific cells and indistinct corpuscles are reexamined after the classification of all samples.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides an image II & SS generation device with a special cell, etc.
After optically enlarging the iIigl and converting it into a two-dimensional electrical signal via a photoelectric conversion element such as an image pickup tube, the Ili processing reproduction scissors device is applied to a cell morphology classification device that distinguishes the cell type. .

Conventionally, a one-person processing device Fi for cell morphology classification such as Jin,
For example, it has been applied to cell diagnostic devices that extract cancer cells from various types of cells, and automatic blood cell classification devices that detect and classify white blood cell heading frequency and abnormal white blood cells.

Recently, these devices have been used to magnify cells and white plum bulbs in #'i specimens using an optical microscope, convert them into analog signals using a photoelectric conversion element tube such as an image pickup tube, and then convert these analog signals into digital signals and store them. At the same time as storing it in the device, the characteristics of cells and white blood cells are extracted and classified based on this digital signal.Only when the classification result is specific, such as unknown cells or abnormal cells, the digital signal is stored in another storage device.
For example, attempts have been made to store classification results and white blood cell names in an external 11A storage device such as a magnetic tape.

However, as the automation of such devices progresses, there is a growing desire to include only specific specimens and visually reexamine them after classification, and such conventional methods require a large amount of memory per cell or white blood cell. As a result, there was a drawback that the number of files that could be stored and stored was small.

In addition, in the method described above, cells in the specimen and leukocytes f#? A method has been disclosed in which after converting into a digital signal, classification is performed, and only when the classification result is a specific cell or an unknown sphere, the signal is converted into a digital signal tube analog signal and stored in a storage device such as a VTR.

However, even with this method, there is a drawback that the resolution of the VTR is low, so the image quality upon re-examination is poor and it is difficult to re-examine. That is, this method first converts the television signal sent to the display device into a digital signal, and then converts it into an analog signal only in the case of specific cells or unknown spheres, stores it in the video track of the VTR, and records the specimen number. The results of the classification and the type of each blood cell are stored in the audio trunk. However, in such a method, when a specific cell or an unknown sphere is determined, the next screen is captured and the image information and discrimination information cannot be stored in synchronization with sound.

In addition, methods have been attempted in which digital images of specific cells and unknown spheres are stored together with signal tube identification information in an external storage device such as magnetic tape. It is necessary to convert and store the television signal tube digital signal of the entire screen when it appears. On the other hand, 1
When converting a TV signal on a screen to a digital code, for example, if the resolution of one 11m surface is equivalent to 0.25μ on the specimen, and if the whole screen is divided into 320 horizontally and 240 vertically, the entire screen will be 80μ horizontally on the specimen. It is divided into 6800 picture elements by taking an image of 60μ vertically.

Furthermore, television signals consist of the primary color red, green, and blue signals, and when each color is converted to an 8-bit digital signal, each signal has 7 bits.
Since the storage capacity is 6800 bytes, the storage capacity t for one entire screen is tripled to 2304007 bytes.

However, in some specimens, a large number of specific cells or unknown spheres appear, and the memory capacity of the above-mentioned 111iii surface is
When you think about it, the number of samples that can be stored in the storage device decreases.

An object of the present invention is to provide a processing and reproducing device that enables reexamination of specific cells and unknown spheres after the classification of all specimens, increases the number of specimens required for reexamination, and improves reliability. It is in.

The present invention confirms that the digital image signal at the time of classification requires only screen information slightly larger than that of white blood cells, and reduces the storage capacity t required to store the digital image signal of specific cells and unknown cells. This is what I did.

Embodiment 1 of the present invention will be explained below with reference to FIGS. 1 to 4. Note that the embodiment described below uses the pattern wt# &
The technical meeting will be used to automatically classify white blood cells in blood images @Iioki meeting.

In FIG. 1, a blood image on a blood smear is optically magnified by an optical microscope 1, and is imaged in gradation on an image sensor 3 having a large number of semiconductor elements arranged in one dimension. Specimen pick up on optical microscope 1, macro computer 1
Scanning is performed by a command from 2 via a digital control circuit 7 and a stage drive circuit 6 built into the microcomputer.

At this time, the blood image on the blood smear forms a light gray area on the image 723, is converted into an analog signal, and is then guided to the white blood cell detection circuit 5.

When white blood cells enter the field of view of the optical microscope 1, the digital control circuit 7 and the stage drive circuit 6 are controlled by the command from the white blood cell detection circuit 5 to stop them at the center of the field of view of the optical microscope 1. Auto focus is taken.

A white blood cell image at this time is captured by a television camera 2 including a control unit. The analog signal from the television camera 2 is guided to an analog-to-digital conversion circuit 4 and a multiplexer circuit 13. The output signal of the multiplexer circuit 13 is guided to the multiplexer circuit 14 together with the character signal from the microcomputer 12, and the color monitor 15 can switch between a blood image and a character display according to a command from the microcomputer 12. .

The analog-to-digital conversion circuit 4 converts the analog image signal into a digital image signal in response to the synchronization signal from the feature extraction circuit 11. Is this converted signal the feature extraction circuit 11? The data is stored in the memory circuit 9 via the memory circuit 9. Further, the synchronization signal from the feature extraction circuit 11 is supplied to the television camera 2 and the digital-to-analog conversion circuit IO. 1me memory circuit 9
Digital painting gI! The signal is taken into a feature extraction circuit 11, and characteristic sounds required for classification are obtained. At this time, if the classification result is an unknown ball or an abnormal ball, the microcomputer 12 stores the digital image signal of the LLI memory circuit 9 together with the specimen honor code and discrimination information in an external storage device 8 using, for example, a magnetic tape. At the same time, the next white blood cell detection command tube digital control circuit 7 is issued.

When the classification of the required number of specimens is completed in this manner, the results are outputted to an output device 16 such as a printer.

When performing sample re-examination 1 after classification is completed, the digital 111g1 signal is read out from the external storage device 8 in response to a command from the microcomputer 12, and the Fmfll storage circuit 9, digital-to-analog conversion circuit 10, and output multiplexer circuit 13.14 are read out from the external storage device 8. Can be switched.

The analog signal from the TV camera 2 mentioned above is analog -
The synchronization signal from the feature extraction circuit 11 is exchanged into a digital image signal via the digital conversion circuit 4, and the feature extraction circuit is completed! As shown in FIG. 3, the 0% fine extraction circuit 11 guided by the 0% fine extraction circuit 11 first finds the center position of white blood cells from the characteristic sound of the concentration distribution of the entire television screen 29.

On the other hand, a feature extraction visual field 30 that is slightly wider than the white blood cell image is set at the center of this white blood cell. Feature extraction field of view 30
When is set, the analog signal of the image extraction field of view 30 is converted into a digital image signal via the analog-to-digital conversion circuit 4 based on the synchronization signal from the feature extraction circuit 11, and is stored in the image storage circuit 9. Ru. The feature extraction circuit 11 extracts the number of white blood cells based on the digital image signal of the single storage circuit 9.
eLet's do it. If the result is an unknown ball or an abnormal ball, use external storage device a.
18 stores digital image signals from the image memory circuit 9 according to instructions from the microcomputer 12.

Furthermore, an external storage device 8 and a digital-to-analog conversion circuit 10i1 are configured as shown in FIG. 2 to enable reexamination of specimens such as unknown balls and abnormal balls. In FIG. 2, the one-time storage circuit 9 consists of a high-speed IC memory circuit 17 and output buffers 18 and 19. During the sorting operation, the buffer 19 is
is made conductive to supply a digital image signal to the seven feature extraction circuit 11. On the other hand, when re-examining the specimen, the buffer 181 is turned on and the digital image signal from the external storage device 8 is supplied to the digital-to-analog conversion circuit IO.

The digital image signal from the external storage device 8 is synchronized with the vertical synchronization signal from the feature extraction circuit 11 and is processed by the microcomputer 12 during the vertical blanking period.
The information is read into the internal storage circuit 9.

The digital aij# signal from the buffer 18 is synchronized with the horizontal synchronization signal from the feature extraction port 1811 and is sent to a high-speed digital-to-analog converter 2G for the three primary color red, green, and blue signals.
, 21 and 22, the signal is converted into an analog image signal and guided to analog reproduction circuits 26, 27 and 28. In the analog reproduction circuits 26, 27 and 28, a reproduction synchronization circuit 2 generates a reproduction synchronization signal based on the synchronization signal of the feature extraction circuit 11.
5, an analog image signal tube for displaying only the feature extraction field of view 30 is synthesized and supplied to the analog switch 23. An analog signal from the television camera 2 is supplied to the analog switch 24, and a white blood cell image for reexamination is displayed on the color monitor 15 according to a command from the microcomputer 12.

As is clear from the above description, according to the present invention, the following young fruit can be obtained. That is, in the present invention, when an unknown cell or an abnormal cell appears, the digital image signal of only the vicinity of the white blood cell is stored, so there is an advantage that the storage capacity per white blood cell can be reduced.

That is, since only a 32μ horizontal by 32μ vertical screen near the white blood cells is displayed, the display screen is divided into 16,384 picture elements. Therefore, the storage capacity for each white blood cell is only 49,152 bytes.

can be significantly reduced. As a result, the digital 1 signal can be efficiently stored in the external storage device.

Although the embodiment described above is applied to an automatic blood cell sorting device for classifying white blood cells, the present invention can also be applied to a cell diagnostic device for detecting cancer cells.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an automatic blood image separation* apparatus to which the present invention is fully applied, Fig. 2 is a block diagram of the main parts of the present invention, Fig. 3 is a detailed explanatory diagram of the present invention, and Fig. 4 is the principle of operation. 1 is a timing diagram for explaining FIG.

Claims (1)

[Claims] 1. A photoelectric conversion means that converts the two-dimensional optical signal of the object to be measured into an analog signal while attaching the sample tube to be measured and moving it under the optical microscope, and converting the analog signal from this photoelectric conversion means into a digital signal. An analog-to-digital conversion means for converting, a first storage means for two-dimensionally storing the digital information from the analog-to-digital conversion means, and an object to be measured based on the digital number 1 from the first storage means. A means for extracting a percent characteristic of an object, and a means for storing digital information of only a specific object to be measured among the digital information in a first storage means based on the result from the feature extracting means.
2 storage means, and a means for reproducing the digital signal fr of a specific object to be measured stored in the storage means #I2; Among the digital signals from the means, means for storing only the digital signals near the object to be measured in the first storage means, and converting the digital signals extracted from the second storage means into analog signals for reproducing the digital signal tube. A digital-to-analog conversion means is provided, and the analog signal for reproducing only the object to be measured from the digital-to-analog conversion means is converted into one screen of analog signals based on the synchronization signal from the feature extraction means, and then displayed. 1. Keyaki processing and regeneration equipment.