阅读说明：本技术 一种骨髓细胞形态学自动检测扫描结构及扫描方法 (Bone marrow cell morphology automatic detection scanning structure and scanning method ) 是由 杨武晨 彭贤贵 张曦 张�诚 张洪洋 陶廷露 张云 苟阳 邓小娟 于 2020-05-25 设计创作，主要内容包括：本发明创造属于骨髓细胞形态学检测领域,具体涉及了一种骨髓细胞形态学自动检测扫描结构及扫描方法。为了解决现有扫描装置不适合对涂片进行扫描质检的问题,本发明提供一种骨髓细胞形态学自动检测扫描结构,包括电源模块,其特征在于,还包括传动装置、光源盒、扫描头和中心处理器,光源盒内安装有光源。光源电连接。传动装置安装在光源盒上,与光源盒转动连接,与电源模块电连接。扫描头安装在光源盒上方,与中心处理器电连接。传动装置与中心处理器电连接。(The invention belongs to the field of bone marrow cell morphology detection, and particularly relates to an automatic bone marrow cell morphology detection scanning structure and a scanning method. In order to solve the problem that the existing scanning device is not suitable for scanning quality inspection of smears, the invention provides an automatic bone marrow cell morphology detection scanning structure which comprises a power supply module and is characterized by further comprising a transmission device, a light source box, a scanning head and a central processor, wherein a light source is arranged in the light source box. The light source is electrically connected. The transmission device is installed on the light source box, is connected with the light source box in a rotating mode and is electrically connected with the power supply module. The scanning head is arranged above the light source box and is electrically connected with the central processor. The transmission device is electrically connected with the central processor.)

1. The automatic bone marrow cell morphology detecting and scanning structure includes power source module, and features that the automatic bone marrow cell morphology detecting and scanning structure includes also driving unit, light source box with light source inside the light source box; the light source is electrically connected; the transmission device is arranged on the light source box, is rotatably connected with the light source box and is electrically connected with the power supply module; the scanning head is arranged above the light source box and is electrically connected with the central processor; the transmission device is electrically connected with the central processor.

2. The automatic bone marrow cell morphology detecting and scanning structure as claimed in claim 1, wherein the light source box is uncovered from top to bottom, and a plurality of rollers are installed at both ends, and the rollers are rotatably connected with the light source box.

3. The bone marrow cell morphology detection scanning structure according to claim 2, characterized in that said transmission device comprises a transmission motor and a transmission belt; the transmission motor and the roller wheel are in coaxial transmission; the transmission motor is electrically connected with the central processor; the transmission belt is sleeved on the roller wheel, and the inner surface of the transmission belt is contacted with the roller wheel; the conveying belt is provided with a smear mounting groove for placing smears.

4. The scanning structure of automatic detection of bone marrow cell morphology as claimed in claim 1, wherein said scanning head is composed of several lens groups; the lens group comprises a plurality of optical microscope lenses and an image sensor; each lens group is provided with at least two rows of microscope lenses which are arranged in a staggered manner; an image sensor is arranged behind each microscope lens; the image sensor is electrically connected with the central processor; the magnification of each lens group is different.

5. An automatic bone marrow cell morphology detecting and scanning method, which is suitable for an automatic bone marrow cell morphology detecting and scanning structure according to claims 1-4, and comprises the following steps:

a1: starting a transmission device, a light source box and an image acquisition module to scan and acquire the smear;

a2: grouping the scanned images according to the magnification, establishing image groups, identifying whether the images in each image group are successfully collected, if so, skipping to A3, and if not, skipping to A1;

a3: establishing corresponding background image frames under each multiple, dividing the image frames into regions according to the regions which are responsible for acquisition of each image sensor in the group, and marking the regions;

a4: the method comprises the steps of backing up pictures acquired by an image acquisition module, placing the pictures into corresponding image frames according to corresponding groups and serial numbers for integration, and stacking and covering overlapped parts;

a5: detecting whether the integrated images have offset, if the images have no offset, jumping to A6, and if the images have offset, jumping to A7;

a6: splicing the images together and uploading;

a7: adjust the image position so that there is no offset and jump to a 6.

6. The method for automatically detecting and scanning bone marrow cell morphology according to claim 5, wherein said A2 comprises the following steps:

b1: identifying whether an observable target is present in each image;

b2: if no observable target exists, jump B5; if an observable target exists, jump to B3;

b3: identifying whether the image has a residual shadow, and if so, jumping to B5; if no ghost exists, jumping to B4;

b4: carrying out boundary identification, if the boundary can not be clearly identified, jumping to B6, and if the boundary can be clearly identified, jumping to B8;

b5: detecting the scanning times, if the scanning times n is less than 3, scanning the smear again, recording the scanning times as n +1, and jumping to B1; if the scanning time n > is 3, jumping to B7;

b6: adjusting the brightness and contrast of the picture and jumping to B4;

b7: stopping working and sending out an error notice to remind a worker to overhaul the equipment or the smear is an invalid smear;

b8: if it is clearly identifiable, step A3 is performed.

7. The method for automatically detecting and scanning bone marrow cell morphology according to claim 5, wherein said A4 comprises the following steps:

c1: identifying a starting end and a terminal end of each image;

c2: aligning the end point of each image, then placing the two images positioned at the two sides to the corresponding positions of the image frame, and enabling the sides of the two images to be superposed with the sides of the image frame;

c3: and then, sequentially placing adjacent images from two sides inwards to corresponding intervals, overlapping overlapped areas, and finally integrating a plurality of images into a whole image.

8. The method for automatically detecting and scanning bone marrow cell morphology according to claim 5, wherein said A5 comprises the following steps:

d1: intercepting the images at the overlapping position and the periphery of each adjacent interval, and numbering according to the image sequence number;

d2: searching whether an adjacent breakpoint exists on the edge line of the overlapped position of the intercepted picture, and if so, skipping to D3; if not, jump A6;

d3: if the number of the breakpoints existing on the edge line is more than or equal to 4, the integration of the pictures is deviated, and jumping to A7; if the number of the break points is less than 4, jump A6.

9. The automatic bone marrow cell morphology detecting and scanning method according to claim 8, characterized in that the set of images created is named as R, said R is a positive integer; the sequence number of the image in each group is named as Q, and Q belongs to (1, m); m is a positive integer, said Q₁＝1，Q_mM; the number of the superposition position is named as K, and the K belongs to (2, m); said K₁＝2，K_m＝m。

10. The method for automatically detecting and scanning bone marrow cell morphology according to claim 9, wherein said A7 comprises the following steps:

e1: acquiring screenshot numbers at overlapped positions with offsets, and sequencing the numbers;

e2: let the first number of the sequence number be K_m-aThe last number is K_m-bThat isTaking the No. 1 image to the No. m-a-1 image as a whole and naming the whole as the No. 1 block; taking the m-b images to the m images as a whole, and naming the whole as a No. 2 block; taking the images from m-a to m-b-1 as a whole, and naming the whole as a No. 3 block; move block 2, eliminate at K_m-aIs offset or occurs at K_m-bOr eliminated simultaneously; move block 2, eliminate at K_m-aIs offset or occurs at K_m-bOr eliminated simultaneously;

e3: the number of overlapping regions of the image where the drift is eliminated is shifted out of the ordering and the step E2 is repeated until no shift occurs.

Technical Field

The invention belongs to the field of bone marrow cell morphology detection, and particularly relates to an automatic bone marrow cell morphology detection scanning structure and a scanning method.

Background

Myelomorphic microscopy is one of the key diagnostic tools in hematology and is commonly used to diagnose a variety of conditions including leukemia, multiple myeloma, lymphoma, anemia, and pancytopenia. According to the guidelines for diagnosing malignant tumors of the bone marrow issued by the world health organization, detailed and accurate manual microscopic examination is required for diagnosis. Normal bone marrow contains all differentiated cells at developmental stages, from early precursor stem cells to functionally mature cells, including hematopoietic stem cells, which are the precursors of most blood cells, as well as mesenchymal and endothelial stem cells, which are considered to be the gatekeeper cells of the bone marrow. The morphological characteristics of these cells depend on their own biological characteristics and are affected by the process of smear, staining and image acquisition.

The quality inspection of the quality of the bone marrow cell smear and the acquisition and analysis of image information are very critical, and the accuracy of the later analysis result is directly determined. Therefore, it is important to solve the quality inspection and image scanning analysis of smears by different dimensional technical means.

Disclosure of Invention

In order to solve the problem that the existing scanning device is not suitable for scanning the smear, the invention provides a scanning structure and a scanning method of an automatic detection system suitable for scanning a bone marrow cell smear.

In order to achieve the above object, the invention adopts the technical scheme that the automatic bone marrow cell morphology detection scanning structure comprises a power module, and is characterized by further comprising a transmission device, a light source box, a scanning head and a central processor, wherein a light source is arranged in the light source box; the light source is electrically connected; the transmission device is arranged on the light source box, is rotatably connected with the light source box and is electrically connected with the power supply module; the scanning head is arranged above the light source box and is electrically connected with the central processor; the transmission device is electrically connected with the central processor.

Preferably, the light source box is uncovered from top to bottom, and a plurality of rollers are installed at two ends of the light source box, and the rollers are rotatably connected with the light source box.

Preferably, the transmission device comprises a transmission motor and a transmission belt; the transmission motor and the roller wheel are in coaxial transmission; the transmission motor is electrically connected with the central processor; the transmission belt is sleeved on the roller wheel, and the inner surface of the transmission belt is contacted with the roller wheel; the conveying belt is provided with a smear mounting groove for placing smears. The smear mounting groove is a through groove.

Preferably, the scanning head consists of a plurality of lens groups; the lens group comprises a plurality of optical microscope lenses and an image sensor; each lens group is provided with at least two rows of microscope lenses which are arranged in a staggered manner; an image sensor is arranged behind each microscope lens; the image sensor is electrically connected with the central processor; the magnification of each lens group is different.

Preferably, the automatic detecting and scanning method for bone marrow cell morphology is applicable to the automatic detecting and scanning structure for bone marrow cell morphology, and is characterized by comprising the following steps: a1: starting a transmission device, a light source box and an image acquisition module to scan and acquire the smear; a2: grouping the scanned images according to the magnification, establishing image groups, identifying whether the images in each image group are successfully collected, if so, skipping to A3, and if not, skipping to A1; a3: establishing corresponding background image frames under each multiple, dividing the image frames into regions according to the regions which are responsible for acquisition of each image sensor in the group, and marking the regions; a4: the method comprises the steps of backing up pictures acquired by an image acquisition module, placing the pictures into corresponding image frames according to corresponding groups and serial numbers for integration, and stacking and covering overlapped parts; a5: detecting whether the integrated images have offset, if the images have no offset, jumping to A6, and if the images have offset, jumping to A7; a6: splicing the images together and uploading; a7: adjust the image position so that there is no offset and jump to a 6.

Preferably, the A2 comprises the following steps: b1: identifying whether an observable target is present in each image; b2: if no observable target exists, jump B5; if an observable target exists, jump to B3; b3: identifying whether the image has a residual shadow, and if so, jumping to B5; if no ghost exists, jumping to B4; b4: carrying out boundary identification, if the boundary can not be clearly identified, jumping to B6, and if the boundary can be clearly identified, jumping to B8; b5: detecting the scanning times, if the scanning times n is less than 3, scanning the smear again, recording the scanning times as n +1, and jumping to B1; if the scanning time n > is 3, jumping to B7; b6: adjusting the brightness and contrast of the picture and jumping to B4; b7: stopping working and sending out an error notice to remind a worker to overhaul the equipment or the smear is an invalid smear; b8: if it is clearly identifiable, step A3 is performed.

Preferably, the A4 comprises the following steps: c1: identifying a starting end and a terminal end of each image; c2: aligning the end point of each image, then placing the two images positioned at the two sides to the corresponding positions of the image frame, and enabling the sides of the two images to be superposed with the sides of the image frame; c3: and then, sequentially placing adjacent images from two sides inwards to corresponding intervals, overlapping overlapped areas, and finally integrating a plurality of images into a whole image.

Preferably, the A5 comprises the following steps: d1: intercepting the images at the overlapping position and the periphery of each adjacent interval, and numbering according to the image sequence number; d2: searching whether an adjacent breakpoint exists on the edge line of the overlapped position of the intercepted picture, and if so, skipping to D3; if not, jump A6; d3: if the number of the breakpoints existing on the edge line is more than or equal to 4, the integration of the pictures is deviated, and jumping to A7; if the number of the break points is less than 4, jump A6.

Preferably, the established image group is named as R, and the R is a positive integer; the sequence number of the image in each group is named as Q, and Q belongs to (1, m); m is a positive integer, said Q₁＝1，Q_mM; the number of the superposition position is named as K, and the K belongs to (2, m); said K₁＝2，K_m＝m。

Preferably, the A7 comprises the following steps: e1: acquiring screenshot numbers at overlapped positions with offsets, and sequencing the numbers; e2: let the first number of the sequence number be K_m-aThe last number is K_m-bThen, the image No. 1 to the image No. m-a-1 are regarded as a whole and named as the block No. 1; taking the m-b images to the m images as a whole, and naming the whole as a No. 2 block; taking the images from m-a to m-b-1 as a whole, and naming the whole as a No. 3 block; move block 2, eliminate at K_m-aIs offset or occurs at K_m-bOr eliminated simultaneously; e3: the number of overlapping regions of the image where the drift is eliminated is shifted out of the ordering and the step E2 is repeated until no shift occurs.

The beneficial effects created by the invention are as follows: the quality inspection of the marrow cell smear and the accurate scanning of the image are realized through the optimized setting of the scanning structure and the reasonable matching of the image scanning.

Drawings

FIG. 1: bone marrow cell morphology automatic detection scanning structure overall schematic diagram

FIG. 2: bottom structure of scanning head

FIG. 3: picture processing flow

FIG. 4: exemplary diagrams of Picture Shift processing

In the figure: 1. light source box, 2, running roller, 3, smear mounting groove, 4, transmission band, 5, scanning head, 6, lens assembly.

Detailed Description

As shown in fig. 1 and fig. 2, an automatic bone marrow cell morphology detecting and scanning structure comprises a power supply module, a light source box 1, a scanning head 5 and a central processor.

A light source is arranged in the light source box 1, and the light source 1 is electrically connected with the power supply module. The light source box 1 is provided with a transmission device which is rotationally connected with the light source box 1, and the transmission device is electrically connected with the power supply module and the central processor. The light source box 1 has no cover up and down, and a plurality of rollers 2 are installed at two ends, and the rollers 2 are rotatably connected with the light source box 1.

The moving device comprises a transmission motor and a transmission belt 4. The transmission motor is in coaxial transmission with the roller 2 and is electrically connected with the central processor. The transmission band 4 is sleeved on the roller 2, and the inner surface of the transmission band is contacted with the roller 2. The transmission band 4 is provided with a smear mounting groove 3 for placing smears. The smear mounting groove 3 is a through groove.

The scanning head 5 is installed above the light source box 1 and is electrically connected with the central processor. The scanning head 5 is composed of a plurality of lens groups 6. The lens groups 6 comprise a plurality of optical microscope lenses and image sensors, and each lens group 6 is provided with at least two rows of microscope lenses which are arranged in a staggered way. And an image sensor, an image sensor and a central processor are arranged behind each microscope lens. The magnification is different between each lens group 6.

The working principle of the scanning structure is that the smear is driven to move by the transmission belt 4 and sequentially passes through the lens groups 6 with different magnification factors, and then the light inside the light source box 1 enters the lens from the passing smear and is captured by the image sensor, which is the same as the principle of an optical microscope.

The objective lenses of present microscopes are designed smaller for image clarity. If too much light enters, the brightness of the seen image is too high, and the image in the visual field cannot be distinguished, so that the image is not beneficial to observation, and the whole smear cannot be completely observed at one time through one lens and needs to be continuously converted. So this application is installed the camera lens in row, but again because objective actual size is bigger than the camera lens, so if only single row installation can make again can not carry out complete scanning to the smear, so just be provided with the camera lens of multirow dislocation array and form lens group 6, be provided with the lens group 6 of different magnifications again simultaneously, so can once only obtain the complete scanogram of different magnifications.

However, this method causes a problem that a part of the content is repeatedly scanned, and thus how to stitch a plurality of images having overlapping properties into a complete image is encountered.

As shown in fig. 3, an automatic detecting and scanning method for morphology of bone marrow cells, which is suitable for the above automatic detecting and scanning structure for morphology of bone marrow cells, includes the following steps: a1: and starting the transmission device, the light source box 1 and the image acquisition module to scan and acquire the smear. A2: grouping the scanned images according to the magnification, establishing image groups, identifying whether the images in each image group are successfully acquired, and jumping to A3 if the images in each image group are successfully acquired, or jumping to A1 if the images in each image group are failed. A3: and establishing corresponding background image frames under each multiple, dividing the image frames into regions according to the regions which are acquired by each image sensor in the group, and marking the regions. A4: and backing up the pictures acquired by the image acquisition module, placing the pictures into corresponding image frames according to corresponding groups and serial numbers for integration, and stacking and covering the overlapped parts. A5: detecting whether the integrated image has an offset, jumping to A6 if the image has no offset, and jumping to A7 if the image has an offset. A6: the images are stitched together and uploaded. A7: adjust the image position so that there is no offset and jump to a 6.

A5 comprises the following steps: d1: and intercepting the overlapped position between every two adjacent regions and the pictures around the overlapped position, and numbering according to the image sequence number. D2: and finding whether an adjacent breakpoint exists on an edge line of the overlapped position of the intercepted picture, and if so, jumping to D3. If not, jump A6. D3: if the number of the break points existing on the edge line is more than or equal to 4, the integration of the pictures is indicated to have an offset, and the jump is A7. If the number of the break points is less than 4, jump A6.

The established image group is named as R, and the R is a positive integer. The sequence number of the image in each group is named as Q, and Q belongs to (1, m). m is a positive integer, said Q₁＝1，Q_mM. The number of the coincidence position is named as K, and the K belongs to (2, m). Said K₁＝2，K_m＝m。

A2 comprises the following steps: b1: the presence or absence of observable objects in each image is identified. B2: if no observable target exists, jump B5. If there is an observable target, jump B3. B3: and identifying whether the image has the residual shadow or not, and if so, jumping to B5. If there is no ghost, jump to B4. B4: boundary identification is performed, and if the boundary cannot be clearly identified, a jump is made to B6, and if the boundary can be clearly identified, a jump is made to B8. B5: the number of scans is detected, if n <3, the smear is rescanned and recorded as n +1, and B1 is skipped. If the scanning time n > is 3, jump B7. B6: adjust the brightness and contrast of the picture and jump to B4. B7: stopping working and sending out an error notice to remind a worker to overhaul the equipment or the smear is an invalid smear. B8: if it is clearly identifiable, step A3 is performed.

A2 is a quality inspection process for images, which can eliminate some scanning structures generated by mechanical failure or thought misoperation for screening, and can also judge the problems of the scanning structures according to the quality inspection result.

In this application, scanning structure is when independently examining quality, if the picture highlight has all appeared and has not observed the content for three consecutive times, can inform the operator and not place the smear or placed blank smear, has appeared wherein that other phenomena are mixed, explains this device and has appeared the problem, if all appear the ghost image problem in the cubic scanning, explains that the moving speed of the transmission band 4 of structure is too fast or transmission band 4 has appeared the shake. If the picture brightness problem appears for many times, the brightness of the structure is over high.

The judgment of the ghost image is to detect the width and the coincidence degree of the lines, and if the central processor finds that the lines are thick and meaningless lines exist, the ghost image is judged to exist.

As shown in fig. 4, a4 includes the following steps: c1: the beginning and end of each image are identified. C2: and aligning the end points of each image, and then placing the two images positioned at the two sides to the corresponding positions of the image frame, so that the sides of the two images are superposed with the sides of the image frame. C3: and then, sequentially placing adjacent images from two sides inwards to corresponding intervals, overlapping overlapped areas, and finally integrating a plurality of images into a whole image.

A7 comprises the following steps: e1: and acquiring the screenshot numbers at the overlapped positions with the offset, and sequencing the numbers. E2: let the first number of the sequence number be K_m-aThe last number is K_m-bThen image # 1 through m-a-1 are considered as a whole and named block # 1. The m-b image to the m image are regarded as a whole and named as the No. 2 block. The m-a image to the m-b-1 image are regarded as a whole and named as a No. 3 block. Move block 2, eliminate at K_m-aIs offset or occurs at K_m-bOr both. E3: the number of overlapping regions of the image where the drift is eliminated is shifted out of the ordering and the step E2 is repeated until no shift occurs.

In fig. 4, m is 6, so Q is named 1 to 6, K is named 2 to 6, and for the sake of distinction, the image numbers in each image group are represented by (r), (c), and (r) in the drawing, and the number of each overlapping area is represented by II III IV V VI. So if m-a is 3, m-b is 5. If the overlapped part with the number IV is also offset, the image I and the image II are regarded as a whole and named as a block I, the image I and the image II are regarded as a whole and named as a block 3, the image III and the image II are regarded as a whole and named as a block 2, and then the block II is moved to eliminate the offset of the overlapped part III or the offset of the overlapped part V. Then, checking is carried out, blocks are divided again, and offset of the IV number overlapping part and the overlapping part corresponding to the IV number overlapping part is eliminated.

The reason why the scanning structure of the present application adopts the imaging mode of the optical microscope to perform scanning instead of the mode of the electron microscope is that the optical microscope is a commonly used observation means at present, so that a lot of reference data can be provided, and the lack of reference data for cell identification due to different imaging modes can be avoided.

The invention adopts an image splicing mode of three fixed positions, two moving positions and one moving position, greatly improves the image splicing efficiency, ensures the image quality and saves the memory occupation of the central processor.

In the scanning process, due to the existence of a plurality of lens groups 6, pictures with different magnifications are formed, and the magnifications of the lens groups 6 are 100, 200, 400 and 1000 times from small to large; of course, the formed picture can still be subjected to isometric amplification.

In the structure, one end of the light source box 1 can be positioned outside the integral structure to form an operation platform for facilitating the taking and placing of the smear; smear mounting groove 3 can move to operation platform when using on, the user places the smear in smear mounting groove 3, then starts the structure for the motor drives running roller 2 and rotates, thereby drive the transmission of transmission band 4, transport the smear to the other end, and scan, after moving the other end, motor reverse rotation accomplishes the secondary scanning, forms standby data, then the smear mounting groove reappears on the operation panel, be used for the operator to take out the smear.

The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations can be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the concepts of the present invention are all within the scope of protection defined by the claims.