CN112432902A - Automatic detection system and method for judging cell number through peripheral blood cell morphology - Google Patents
Automatic detection system and method for judging cell number through peripheral blood cell morphology Download PDFInfo
- Publication number
- CN112432902A CN112432902A CN202011395584.9A CN202011395584A CN112432902A CN 112432902 A CN112432902 A CN 112432902A CN 202011395584 A CN202011395584 A CN 202011395584A CN 112432902 A CN112432902 A CN 112432902A
- Authority
- CN
- China
- Prior art keywords
- cells
- area
- peripheral blood
- image acquisition
- cell
- 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
- 210000004027 cell Anatomy 0.000 title claims abstract description 154
- 210000004976 peripheral blood cell Anatomy 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 238000004891 communication Methods 0.000 claims abstract description 33
- 230000003993 interaction Effects 0.000 claims abstract description 12
- 230000007246 mechanism Effects 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 7
- 230000000877 morphologic effect Effects 0.000 claims description 31
- 210000003651 basophil Anatomy 0.000 claims description 27
- 210000003979 eosinophil Anatomy 0.000 claims description 27
- 210000003743 erythrocyte Anatomy 0.000 claims description 27
- 210000004698 lymphocyte Anatomy 0.000 claims description 27
- 210000001616 monocyte Anatomy 0.000 claims description 27
- 210000000440 neutrophil Anatomy 0.000 claims description 27
- 210000001995 reticulocyte Anatomy 0.000 claims description 27
- 230000002159 abnormal effect Effects 0.000 claims description 23
- 210000001772 blood platelet Anatomy 0.000 claims description 22
- 238000013527 convolutional neural network Methods 0.000 claims description 19
- 238000005192 partition Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 7
- 208000007502 anemia Diseases 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 208000032839 leukemia Diseases 0.000 claims description 7
- 210000000265 leukocyte Anatomy 0.000 claims description 7
- 230000002071 myeloproliferative effect Effects 0.000 claims description 7
- 208000011580 syndromic disease Diseases 0.000 claims description 7
- 230000001575 pathological effect Effects 0.000 claims description 6
- 210000000601 blood cell Anatomy 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 claims 1
- 201000010099 disease Diseases 0.000 abstract description 7
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 7
- 238000003745 diagnosis Methods 0.000 abstract description 5
- 238000001914 filtration Methods 0.000 abstract description 5
- 238000003384 imaging method Methods 0.000 description 4
- 210000004369 blood Anatomy 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 208000014951 hematologic disease Diseases 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003583 cytomorphological effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004043 dyeing Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000000386 microscopy Methods 0.000 description 1
- 244000045947 parasite Species 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 210000005259 peripheral blood Anatomy 0.000 description 1
- 239000011886 peripheral blood Substances 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/85—Investigating moving fluids or granular solids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention belongs to the technical field of peripheral blood cell morphology detection, and particularly relates to an automatic detection system and method for judging the number of cells by peripheral blood cell morphology, which comprises a main control module, a scanning module and a human-computer interaction mechanism, wherein the main control module is in communication connection with the scanning module, the human-computer interaction mechanism is in communication connection with the main control module, the scanning module comprises a light source device, a filtering device and an image acquisition device, the filtering device is loaded on the image acquisition device, the image acquisition device is in communication connection with the main control module, the human-computer interaction mechanism comprises a computer, a display screen, a warning module and a printer, the main control module is in communication connection with the computer, the main control module is in communication connection with the display screen, the warning module is in communication connection with the display screen, the printer is in electrical connection with the computer, and is used for solving the problem that the workload is large when, the repeatability is poor, time and labor are consumed, and the method is not suitable for screening large-batch specimens and is easy to influence the disease diagnosis.
Description
Technical Field
The invention belongs to the technical field of peripheral blood cell morphology detection, and particularly relates to an automatic detection system and method for judging the number of cells through peripheral blood cell morphology.
Background
Peripheral blood cell detection is a routine task in clinical tests and is of great significance in diagnosis and identification of many diseases, particularly blood diseases. At present, a laboratory generally adopts a blood analyzer to perform classified counting of peripheral blood cells, the blood analyzer generally adopts technologies such as physics and cytochemistry to perform classified counting of cells, the automation degree is higher, but the accuracy is not high. Thus, microscopic examination of peripheral blood remains a necessary procedure for diagnosing many diseases and for performing instrument performance assessment.
The manual microscope microscopic examination is to perform preliminary detection through a low-power microscope, then perform oil microscopic examination and perform manual counting and identification. When manual operation is adopted, the inspection workload is large, the repeatability is poor, time and labor are consumed, and the method is not suitable for screening large-batch specimens; moreover, the objectivity of the judgment result of the doctor is insufficient due to the influence of the experience and the number of the work of the cytomorphological examination worker, and the diagnosis of the disease condition is affected.
Disclosure of Invention
The purpose of the invention is: the automatic detection system and the method for judging the cell number in the peripheral blood cell morphology are provided, and are used for solving the problems of large workload, poor repeatability, time and labor consumption, inapplicability to screening of large-batch samples and easiness in influencing disease diagnosis in the process of manually detecting the peripheral blood cell morphology by microscopy.
In order to achieve the technical purpose, the technical scheme adopted by the invention is as follows:
the utility model provides an automatic check out system of cell figure is differentiateed to peripheral blood cell morphology, includes host system, scanning module and human-computer interaction mechanism, host system and scanning module communication connection, human-computer interaction mechanism and host system communication connection, scanning module includes light source device, filter device and image acquisition device, the filter device loading is on image acquisition device, image acquisition device and host system communication connection, human-computer interaction mechanism includes computer, display screen, warning module and printer, host system and computer communication connection, host system and display screen communication connection, warning module and display screen communication connection, the printer is connected with the computer electricity.
Furthermore, the image acquisition device comprises a camera, an image acquisition module and a compensation module, wherein the camera consists of a plurality of lens groups with different magnification factors; every the lens group comprises the same and misplaced camera lens of a plurality of magnification, image acquisition module comprises a plurality of image sensor, image sensor and host system communication connection, image sensor installs at every camera lens at the back, filter installs on every camera lens, compensation module and image sensor communication connection, compensation module and host system communication connection.
Further, the light source device is a shadowless light source, and the shadowless light source is electrically connected with the main control module.
Further, the automatic detection method for distinguishing the number of the cells by the morphology of the peripheral blood cells comprises the following steps: s1: constructing a morphological database of peripheral blood cells; s2: scanning the whole peripheral blood cells to obtain a hypo-multiple image area, partitioning the hypo-multiple image area, and counting the number of partitions; s3: scanning each subarea with high-power images, and identifying and counting cells in the subareas; s4: carrying out classification statistics on the cells in different partitions, and carrying out weighted calculation on the cells in different partitions to obtain calculation data; s5: comparing the calculated data with normal data for reference, and performing reference analysis; s6: and scanning and judging the abnormal information again.
Further defined, the database of morphology of peripheral blood cells in S1 includes size ranges, various morphologies and numbers of erythrocytes and reticulocytes in normal human blood cells, cell size ranges of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, various morphologies and numbers of cells, size ranges, various morphologies and numbers of platelets; the size range, various forms and number of erythrocytes and reticulocytes in peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndrome and the like in each stage, the cell size range, various forms and number of cells of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, and the size range, various forms and number of platelets.
Further limiting, the partition of the low-magnification image in S2 includes a positive counting area and an expanded counting area, where the expanded counting area is located at the periphery of the positive counting area and is an extension of the positive counting area; the area of the positive counting area is the average area size of a cells; the distance between the boundary of the expanded counting area and the boundary of the positive counting area is the average diameter distance of b cells; when counting cells in the positive counting area, if the cells falling into the positive counting area are found to be incomplete, the cells are extended to the expanded counting area, then the area of the cells in the positive counting area is measured and compared with the area in the expanded counting area, the cells in the two areas are counted if the area in the positive counting area is large, and the cells are not counted if the area in the positive counting area is large.
Further defined, the average area of the cells is the average area of red blood cells, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes, and platelets at the detection site.
Further defined, the step of scanning the high power image in S3 includes the following steps: constructing an 8-layer convolutional neural network, performing shape recognition training on erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets in normal human peripheral blood cells through the 8-layer convolutional neural network, and connecting training data with a morphological database of peripheral blood cells in S1; b: recognizing and training the forms of red blood cells, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and pathological cells in peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndromes and the like in each stage through a convolutional neural network, and connecting training data with a peripheral blood cell morphological database in S1; c: constructing an image acquisition database, dividing the image acquisition database into 9 areas, wherein 8 areas respectively correspond to erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets, and the other area corresponds to the cell morphology which does not appear or is recorded in human peripheral blood cells, extracting the morphological characteristics of the erythrocytes, the reticulocytes, the neutrophils, the eosinophils, the basophils, the monocytes, the lymphocytes and the platelets through image acquisition, screening and combining the same characteristics, and then respectively classifying and counting according to the corresponding image acquisition areas; d: and connecting the image acquisition database with a convolutional neural network, performing the first step of judgment, connecting the image acquisition database with the morphological database of the peripheral blood cells in S1 after the first step of judgment, and further performing the second step of judgment.
In the step C, the image is acquired by color imaging and nine-grid feature locking to extract morphological features of the cells, and by color imaging, the forms of erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets in peripheral blood cells are clearly displayed to avoid cell overlapping or attachment, and by nine-grid feature locking, one cell is divided into different regions, and then the features of the regions are combined into complete cell features, so that morphological feature changes of the cells can be observed.
Further, in the step D, the first determination rule is: and C, comparing the features acquired in the step C with the cell morphology in the convolutional neural network, namely comparing the similarity between different morphologies of the normal cell and the morphology of the detected cell, wherein if the similarity is greater than the dissimilarity, the cell is the normal cell, and if the dissimilarity is greater than the similarity, the second step of judgment is carried out, and the second step of judgment rule is as follows: comparing the cells with dissimilarity degree greater than similarity degree with the cell database in S1, wherein the cells with dissimilarity degree greater than similarity degree are more similar than the cell data in S1, and are normal cells, otherwise, the cells are pathological cells.
The invention has the beneficial effects that: the system constructs a morphological database of peripheral blood cells, statistically classifies various cell morphologies in the peripheral blood cells of normal people, statistically classifies various cell morphologies in the peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndromes and the like at each stage, enables the database to contain all the morphology and counting statistics of the peripheral blood cells, then performs image acquisition and convolutional neural network identification, and accurately identifies, classifies, statistically and summarizes the peripheral blood cells to be detected according to morphology, thereby completing automatic detection of the peripheral blood cell morphology, and avoiding the problems of large workload, poor repeatability, time and labor consumption, inapplicability to screening of large-scale samples and easiness in influencing diagnosis when the peripheral blood cell morphology is detected by manual microscopic examination.
Drawings
The invention is further illustrated by the non-limiting examples given in the accompanying drawings;
FIG. 1 is a schematic diagram of the overall working process of an embodiment of the automatic detection system and method for determining the number of cells by peripheral blood cell morphology according to the present invention;
Detailed Description
The present invention will be described in detail with reference to the drawings and specific embodiments, wherein like reference numerals are used for similar or identical parts in the drawings or the description, and implementations not shown or described in the drawings are known to those of ordinary skill in the art. In addition, directional terms, such as "upper", "lower", "top", "bottom", "left", "right", "front", "rear", and the like, used in the embodiments are only directions referring to the drawings, and are not intended to limit the scope of the present invention.
As shown in fig. 1, the automatic detection system and method for determining the number of cells by peripheral blood cell morphology of the present invention includes a main control module, a scanning module and a human-computer interaction mechanism, wherein the main control module is in communication connection with the scanning module, the human-computer interaction mechanism is in communication connection with the main control module, the scanning module includes a light source device, a light filtering device and an image acquisition device, the light source device is a shadowless light source, and the shadowless light source is electrically connected with the main control module to avoid additional interference generated by the light source. The filtering device is loaded on the image acquisition device, the image acquisition device is in communication connection with the main control module, the human-computer interaction mechanism comprises a computer, a display screen, a warning module and a printer, the main control module is in communication connection with the computer, the main control module is in communication connection with the display screen, the warning module is in communication connection with the display screen, and the printer is in electrical connection with the computer.
The morphological characteristics of peripheral blood cells to be detected are shot through an image acquisition device, the morphological characteristics are transmitted to a main control module, the detected total quantity of erythrocytes, reticulocytes and neutrophils, eosinophils, basophils, monocytes and lymphocytes in the peripheral blood cells is counted and displayed on a display screen through analysis and statistics of the main control module, an area with abnormal display is reminded through a warning module, then the image acquisition device is controlled by a doctor to perform image scanning on the abnormal display area again and display the abnormal display area on the display screen, the abnormal area on the display screen is amplified for manual review, and after the abnormal area is determined, the final scanning result is printed through an operation computer, so that the doctor can directly check the abnormal area.
The image acquisition device comprises a camera, an image acquisition module and a compensation module, wherein the camera consists of a plurality of lens groups with different magnification factors. Each lens group consists of a plurality of lenses with the same magnification and placed in a staggered manner, the image acquisition module consists of a plurality of image sensors, the image sensors are in communication connection with the main control module, the image sensors are installed behind the lenses, the light filtering device is installed on each lens, the compensation module is in communication connection with the image sensors, the compensation module is in communication connection with the main control module, three-step scanning can be carried out when peripheral blood cell scanning is carried out through the lenses with different magnifications, low-power scanning is carried out when first-step scanning is carried out, preliminary scanning is carried out on peripheral blood cell smears, whether the peripheral blood cell smears meet the detection requirements or not is determined, whether dyeing on the smears meets the requirements or not and whether parasites exist or not, after the first-step detection is completed, second-step scanning is carried out, high-power scanning is adopted when second-step scanning is carried out, red blood cells in the, Morphological characteristics of neutrophils, eosinophils, basophils, monocytes and lymphocytes in the reticulocytes and the leukocytes are all scanned out, the overall cell number is counted through analysis and statistics of the main control module, and then normal statistics of various cell numbers and statistics and classification of various morphologically abnormal cell numbers are carried out, so that the result can be displayed on a display screen in a branch pipe. When abnormal reminding exists, the third step of ultrahigh-power scanning is carried out, pixel compensation is carried out on the abnormal area through the compensation module, the cell morphology of the abnormal area can be scanned and extracted more fully and displayed on the display screen, and whether the cell morphology of the abnormal area is abnormal or not is judged by a doctor, so that the scanning result is accurate.
The automatic detection method for distinguishing the number of cells by the morphology of the peripheral blood cells comprises the following steps:
s1: constructing a morphological database of peripheral blood cells; s2: scanning the whole peripheral blood cells to obtain a hypo-multiple image area, partitioning the hypo-multiple image area, and counting the number of partitions; s3: scanning each subarea with high-power images, and identifying and counting cells in the subareas; s4: carrying out classification statistics on the cells in different partitions, and carrying out weighted calculation on the cells in different partitions to obtain calculation data; s5: comparing the calculated data with normal data for reference, and performing reference analysis; s6: and scanning and judging the abnormal information again.
By constructing a morphological database of peripheral blood cells, after dividing the hypo-image area into regions, each region is scanned with high power to identify all cells and cell morphology within the region, then the weighted statistical calculation is carried out on the cells and the cell forms of all the subareas to obtain the number of different cells, and comparing and judging with the cells in the database to obtain the number of normal cells and the number of various abnormal cells, and comparing the abnormal cells with known diseased cells for analysis to determine whether the disease is diseased, and when comparing with the known database, it can not be determined whether the disease is diseased, then the cells in the abnormal area are again subjected to image acquisition and feature analysis, whether the pathological changes exist is judged by a doctor, the number of the whole detection cells can be automatically analyzed and counted, abnormal cells can be automatically compared, marked and recorded, and the detection efficiency is greatly improved.
The morphological database of peripheral blood cells in S1 includes the size ranges, various morphologies and numbers of erythrocytes and reticulocytes in normal human blood cells, the cell size ranges of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, the various morphologies and numbers of cells, the size ranges, various morphologies and numbers of platelets; the size range, various forms and quantity of erythrocytes and reticulocytes in peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndrome and the like in each stage, the cell size range, various forms and quantity of cells of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, the size range, various forms and quantity of platelets, and the cell condition can be conveniently determined according to the detected morphological characteristics of various cells.
The partition of the low-magnification image in S2 includes a positive counting area and an extended counting area, and the extended counting area is located at the periphery of the positive counting area and is an extension of the positive counting area. The area of the positive counting region is the average area of a cells, and the average area of the cells is the average area of erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets at the detection site. The distance between the boundary of the expanded counting area and the boundary of the positive counting area is b cell average diameter distances, when the cells in the positive counting area are counted, if the cells falling into the positive counting area are found to be incomplete, the cells are extended to the expanded counting area, then the area of the cells in the positive counting area and the area of the cells in the expanded counting area are measured, the areas in the two areas are compared, if the area in the positive counting area is large, counting is carried out, and otherwise, counting is not carried out.
The high power image scanning in S3 includes the steps of: and A, constructing an 8-layer convolutional neural network, performing shape recognition training on erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets in normal human peripheral blood cells through the 8-layer convolutional neural network, and connecting training data with a peripheral blood cell shape database in S1, so that the convolutional neural network can recognize various shapes of normal erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets, and improve the recognition accuracy.
B: the morphology of red blood cells, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and diseased cells in peripheral blood cells of patients such as anemia, leukemia and myeloproliferative syndromes at each stage is identified and trained by a convolutional neural network, and the training data is connected to a peripheral blood cell morphology database in S1, so that the convolutional neural network can identify whether red blood cells, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets are diseased or not by the morphology of various cells.
C: constructing an image acquisition database, dividing the image acquisition database into 9 areas, wherein 8 areas respectively correspond to erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets, and the other area corresponds to the cell morphology which is not appeared or recorded in human peripheral blood cells, extracting the morphological characteristics of the erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets by image acquisition, namely the morphological characteristics of the cells are extracted by color imaging and nine-lattice feature locking, and the morphologies of the erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets in the peripheral blood cells are displayed clearly by color imaging, avoiding cell overlapping or attaching, dividing a cell into different areas through nine-lattice feature locking, extracting the features of each area, combining the features into complete cell features, observing morphological feature changes of the cell, screening and combining the same features, classifying and counting the cell features according to the corresponding image acquisition areas, communicating the image acquisition database with the image acquisition module and the compensation module, transmitting cell morphological image information acquired by the image acquisition module to the image acquisition database, filing the cell morphological image information in the image acquisition database according to classification, collecting cell morphological data which cannot be classified to the last area, and recording the cell morphological data in the convolutional neural network.
D: connecting the image acquisition database with a convolutional neural network, performing a first step of judgment, after the first step of judgment, connecting the image acquisition database with a morphological database of peripheral blood cells in S1, and further performing a second step of judgment, wherein in the first step of judgment, cell morphologies in 9 regions in the image acquisition database are respectively identified and compared through 8 layers of convolutional neural networks, namely the different morphologies of normal cells are compared with the similarity of the morphology of detected cells, if the similarity is greater than the dissimilarity, the normal cells are obtained, if the dissimilarity is greater than the similarity, the second step of judgment is performed, and the second step of judgment rules are as follows: comparing the cells with dissimilarity degree greater than similarity with the cell database in S1, if the similarity between the cells with dissimilarity degree greater than similarity and the cell data in S1 is greater than dissimilarity degree, the cells are normal cells, otherwise, the cells are pathological cells, and the cell morphological characteristics which cannot be identified are displayed on a display screen to wait for the judgment of a doctor, so that the normal cell morphology and the pathological cell morphology are accurately distinguished, and the time required by manual judgment is reduced.
S5, counting the number of normal cells in the peripheral blood cells, judging whether the number of the normal cells is identical to the content of the normal peripheral blood cells, matching the abnormal cells with the possible blood diseases, listing the possible percentages, and providing the doctors with the reference.
S6, the doctor can operate the computer to scan the abnormal cell shape again, then screen the abnormal cell shape, and display it on the screen to wait for the doctor' S judgment, so that the doctor only needs to judge a few cell shapes, and the doctor does not need to count the number of various cells in the peripheral blood cells, the system can count the number of normal cells and abnormal cells automatically, and parallel relative suspected cases, so the detection of the peripheral blood cells is more convenient.
The present invention provides an automatic detection system and method for determining the number of cells by peripheral blood cell morphology. The description of the specific embodiments is only intended to facilitate an understanding of the method of the invention and its core ideas. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (10)
1. An automatic detection system for peripheral blood cell morphology to determine cell number, characterized in that: including host system, scanning module and human-computer interaction mechanism, host system and scanning module communication connection, human-computer interaction mechanism and host system communication connection, scanning module includes light source device, filter device and image acquisition device, the filter device loading is on image acquisition device, image acquisition device and host system communication connection, human-computer interaction mechanism includes computer, display screen, warning module and printer, host system and computer communication connection, host system and display screen communication connection, warning module and display screen communication connection, printer and computer electrical connection.
2. The system according to claim 1, wherein the system comprises: the image acquisition device comprises a camera, an image acquisition module and a compensation module, wherein the camera consists of a plurality of lens groups with different magnification; every the lens group comprises the same and misplaced camera lens of a plurality of magnification, image acquisition module comprises a plurality of image sensor, image sensor and host system communication connection, image sensor install at every camera lens at the back, filter installs on every camera lens, compensation module and image sensor communication connection, compensation module and host system communication connection.
3. The system according to claim 2, wherein the automatic detection system for distinguishing the number of the cells by the morphology of the peripheral blood cells comprises: the light source device is a shadowless light source, and the shadowless light source is electrically connected with the main control module.
4. The method according to claim 3, wherein the method comprises the steps of: the method comprises the following steps:
s1: constructing a morphological database of peripheral blood cells;
s2: scanning the whole peripheral blood cells to obtain a hypo-multiple image area, partitioning the hypo-multiple image area, and counting the number of partitions;
s3: scanning each subarea with high-power images, and identifying and counting cells in the subareas;
s4: carrying out classification statistics on the cells in different partitions, and carrying out weighted calculation on the cells in different partitions to obtain calculation data;
s5: comparing the calculated data with normal data for reference, and performing reference analysis;
s6: and scanning and judging the abnormal information again.
5. The method according to claim 4, wherein the method comprises the steps of: the morphological database of peripheral blood cells in S1 includes size ranges, various morphologies and numbers of erythrocytes and reticulocytes in normal human blood cells, cell size ranges of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, various morphologies and numbers of cells, size ranges, various morphologies and numbers of platelets; the size range, various forms and number of erythrocytes and reticulocytes in peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndrome and the like in each stage, the cell size range, various forms and number of cells of neutrophils, eosinophils, basophils, monocytes and lymphocytes in leukocytes, and the size range, various forms and number of platelets.
6. The method according to claim 5, wherein the method comprises the steps of: the partition of the low-magnification image in S2 includes a positive counting area and an extended counting area, where the extended counting area is located at the periphery of the positive counting area and is an extension of the positive counting area; the area of the positive counting area is the average area size of a cells; the distance between the boundary of the expanded counting area and the boundary of the positive counting area is the average diameter distance of b cells; when counting cells in the positive counting area, if the cells falling into the positive counting area are found to be incomplete, the cells are extended to the expanded counting area, then the area of the cells in the positive counting area is measured and compared with the area in the expanded counting area, the cells in the two areas are counted if the area in the positive counting area is large, and the cells are not counted if the area in the positive counting area is large.
7. The method according to claim 6, wherein the method comprises the steps of: the average area of the cells is the average area of erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes, and platelets at the detection site.
8. The method according to claim 7, wherein the method comprises the steps of: the high-power image scanning in the step S3 includes the following steps: constructing an 8-layer convolutional neural network, performing shape recognition training on erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets in normal human peripheral blood cells through the 8-layer convolutional neural network, and connecting training data with a morphological database of peripheral blood cells in S1;
b: recognizing and training the forms of red blood cells, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and pathological cells in peripheral blood cells of patients with anemia, leukemia, myeloproliferative syndromes and the like in each stage through a convolutional neural network, and connecting training data with a peripheral blood cell morphological database in S1;
c: constructing an image acquisition database, dividing the image acquisition database into 9 areas, wherein 8 areas respectively correspond to erythrocytes, reticulocytes, neutrophils, eosinophils, basophils, monocytes, lymphocytes and platelets, and the other area corresponds to the cell morphology which does not appear or is recorded in human peripheral blood cells, extracting the morphological characteristics of the erythrocytes, the reticulocytes, the neutrophils, the eosinophils, the basophils, the monocytes, the lymphocytes and the platelets through image acquisition, screening and combining the same characteristics, and then respectively classifying and counting according to the corresponding image acquisition areas;
d: and connecting the image acquisition database with a convolutional neural network, performing the first step of judgment, connecting the image acquisition database with the morphological database of the peripheral blood cells in S1 after the first step of judgment, and further performing the second step of judgment.
9. The method according to claim 8, wherein the method comprises the steps of: and in the step C, the morphological characteristics of the cells are extracted through color development and nine-lattice characteristic locking in image acquisition.
10. The method according to claim 9, wherein the method comprises the steps of: in the step D, the first step determination rule is: and C, comparing the features acquired in the step C with the cell morphology in the convolutional neural network, namely comparing the similarity between different morphologies of the normal cell and the morphology of the detected cell, wherein if the similarity is greater than the dissimilarity, the cell is the normal cell, and if the dissimilarity is greater than the similarity, the second step of judgment is carried out, and the second step of judgment rule is as follows: comparing the cells with dissimilarity degree greater than similarity degree with the cell database in S1, wherein the cells with dissimilarity degree greater than similarity degree are more similar than the cell data in S1, and are normal cells, otherwise, the cells are pathological cells.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011395584.9A CN112432902A (en) | 2020-12-03 | 2020-12-03 | Automatic detection system and method for judging cell number through peripheral blood cell morphology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011395584.9A CN112432902A (en) | 2020-12-03 | 2020-12-03 | Automatic detection system and method for judging cell number through peripheral blood cell morphology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112432902A true CN112432902A (en) | 2021-03-02 |
Family
ID=74691169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011395584.9A Pending CN112432902A (en) | 2020-12-03 | 2020-12-03 | Automatic detection system and method for judging cell number through peripheral blood cell morphology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112432902A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508909A (en) * | 2020-12-03 | 2021-03-16 | 中国人民解放军陆军军医大学第二附属医院 | Disease association method of peripheral blood cell morphology automatic detection system |
| CN113034479A (en) * | 2021-03-31 | 2021-06-25 | 武汉智博见微医疗科技有限公司 | AA. MDS and MA classification method, device and readable storage medium |
| CN115201092A (en) * | 2022-09-08 | 2022-10-18 | 珠海圣美生物诊断技术有限公司 | Method and device for acquiring cell scanning image |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101407759A (en) * | 2008-11-28 | 2009-04-15 | 北京理工大学 | Blood cell analyzer and analysis method |
| CN102920429A (en) * | 2012-10-07 | 2013-02-13 | 吴士明 | Integrated device of mastopathy diagnosis and treatment |
| CN103745210A (en) * | 2014-01-28 | 2014-04-23 | 爱威科技股份有限公司 | Method and device for classifying white blood cells |
| CN106248559A (en) * | 2016-07-14 | 2016-12-21 | 中国计量大学 | A kind of leukocyte five sorting technique based on degree of depth study |
| CN106618506A (en) * | 2017-03-13 | 2017-05-10 | 谢党恩 | Acousto-optical CT three-dimensional imaging system |
| CN107016660A (en) * | 2016-01-28 | 2017-08-04 | 太豪生医股份有限公司 | Pathological change detection method and pathological change detection device of rotary breast image |
| CN111795919A (en) * | 2020-05-25 | 2020-10-20 | 中国人民解放军陆军军医大学第二附属医院 | Bone marrow cell morphology automatic detection system and working method thereof |
| CN111832389A (en) * | 2020-05-25 | 2020-10-27 | 中国人民解放军陆军军医大学第二附属医院 | Counting and analyzing method of bone marrow cell morphology automatic detection system |
-
2020
- 2020-12-03 CN CN202011395584.9A patent/CN112432902A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101407759A (en) * | 2008-11-28 | 2009-04-15 | 北京理工大学 | Blood cell analyzer and analysis method |
| CN102920429A (en) * | 2012-10-07 | 2013-02-13 | 吴士明 | Integrated device of mastopathy diagnosis and treatment |
| CN103745210A (en) * | 2014-01-28 | 2014-04-23 | 爱威科技股份有限公司 | Method and device for classifying white blood cells |
| CN107016660A (en) * | 2016-01-28 | 2017-08-04 | 太豪生医股份有限公司 | Pathological change detection method and pathological change detection device of rotary breast image |
| CN106248559A (en) * | 2016-07-14 | 2016-12-21 | 中国计量大学 | A kind of leukocyte five sorting technique based on degree of depth study |
| CN106618506A (en) * | 2017-03-13 | 2017-05-10 | 谢党恩 | Acousto-optical CT three-dimensional imaging system |
| CN111795919A (en) * | 2020-05-25 | 2020-10-20 | 中国人民解放军陆军军医大学第二附属医院 | Bone marrow cell morphology automatic detection system and working method thereof |
| CN111832389A (en) * | 2020-05-25 | 2020-10-27 | 中国人民解放军陆军军医大学第二附属医院 | Counting and analyzing method of bone marrow cell morphology automatic detection system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508909A (en) * | 2020-12-03 | 2021-03-16 | 中国人民解放军陆军军医大学第二附属医院 | Disease association method of peripheral blood cell morphology automatic detection system |
| CN112508909B (en) * | 2020-12-03 | 2023-08-25 | 中国人民解放军陆军军医大学第二附属医院 | Disease association method of peripheral blood cell morphology automatic detection system |
| CN113034479A (en) * | 2021-03-31 | 2021-06-25 | 武汉智博见微医疗科技有限公司 | AA. MDS and MA classification method, device and readable storage medium |
| CN113034479B (en) * | 2021-03-31 | 2022-06-03 | 武汉智博见微医疗科技有限公司 | AA. MDS and MA classification method, device and readable storage medium |
| CN115201092A (en) * | 2022-09-08 | 2022-10-18 | 珠海圣美生物诊断技术有限公司 | Method and device for acquiring cell scanning image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10677711B2 (en) | Method and apparatus for automated whole blood sample analyses from microscopy images | |
| CN101339185B (en) | Automatic microscopic imager for detecting cast-off cells and detection method | |
| CN112432902A (en) | Automatic detection system and method for judging cell number through peripheral blood cell morphology | |
| US5933519A (en) | Cytological slide scoring apparatus | |
| CA2905730C (en) | System and method for reviewing and analyzing cytological specimens | |
| EP2757372B1 (en) | Device and method for erythrocyte morphology analysis | |
| CN101685060B (en) | Sample imaging apparatus | |
| CN101809589B (en) | Methods and systems for processing biological specimens utilizing multiple wavelengths | |
| Hortinela et al. | Identification of abnormal red blood cells and diagnosing specific types of anemia using image processing and support vector machine | |
| CN111832389B (en) | Counting and analyzing method of bone marrow cell morphology automatic detection system | |
| JP4864709B2 (en) | A system for determining the staining quality of slides using a scatter plot distribution | |
| CN112001315A (en) | Bone marrow cell classification and identification method based on transfer learning and image texture features | |
| KR20010017092A (en) | Method for counting and analyzing morphology of blood cell automatically | |
| CN102297833A (en) | Method for determination of content of various cells in human blood or bone marrow | |
| CN111274949B (en) | Blood disease white blood cell scatter diagram similarity analysis method based on structural analysis | |
| KR101106386B1 (en) | System for classifying slides using scatter plot distributions | |
| CN111795919A (en) | Bone marrow cell morphology automatic detection system and working method thereof | |
| CN113380318B (en) | Artificial intelligence assisted flow cytometry 40CD immunophenotyping detection method and system | |
| CN112481346B (en) | Automatic early warning system and method for detecting abnormal cells in peripheral blood cell morphology | |
| CN111833297B (en) | Disease association method of marrow cell morphology automatic detection system | |
| CN113869124A (en) | Deep learning-based blood cell morphology classification method and system | |
| CN113330311A (en) | Cell image analysis device and sample analysis method | |
| CN114152557B (en) | Image analysis-based blood cell counting method and system | |
| US20240029458A1 (en) | A method for automated determination of platelet count based on microscopic images of peripheral blood smears | |
| RU159002U1 (en) | DEVICE FOR ANALYSIS OF BLOOD CELLS BASED ON THE ASSESSMENT OF CHARACTERISTICS OF STRUCTURAL ELEMENTS OF NUCLEUS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210302 |
|
| RJ01 | Rejection of invention patent application after publication |