CN1553166A - Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method - Google Patents
Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method Download PDFInfo
- Publication number
- CN1553166A CN1553166A CNA2003101116115A CN200310111611A CN1553166A CN 1553166 A CN1553166 A CN 1553166A CN A2003101116115 A CNA2003101116115 A CN A2003101116115A CN 200310111611 A CN200310111611 A CN 200310111611A CN 1553166 A CN1553166 A CN 1553166A
- Authority
- CN
- China
- Prior art keywords
- image
- cell
- array ccd
- area array
- automatic
- 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.)
- Granted
Links
Images
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
An analyser which comprises light source, microscope, display planar array CCD and computer is featured as the follows: a) filtering system, electric-tuning filter controller and planar CCD are disposed on microscope for quick collection of optical spectrum image, computer is used to change central frequency of band-pass; b) a CCD control and data collection system is used to store digital image into computer for image process and analysis; c) a three-dimensional movable control system is used to realize automatic focusing and scanning and d) a computer is used to carry on image composition, image calibration, automatic division and identification.
Description
Technical field
The present invention relates to a kind of micro-multispectral marrow and peripheral blood cells automatic analyzer and method, belong to biologic medical check and analysis instrument and method and technology field, particularly relate to the fast detecting analyser and the method for human blood cell.
Background technology
Blood disease is one of major disease of serious threat human health, different blood diseases, its patient's blood cell has the variation of different matter (comprising form) and amount, there be time-consuming (about 2 hours/example) in conventional marrow (blood) cell Wright's staining artificial discriminator under light microscopic, subjectivity strong (people of the different level of understandings, or same people's different time to observe classification results different).Be unfavorable for the accurate diagnosis and treatment to the patient, therefore replacing the manual sort with intelligentized blood cytoanalyze and method is the important topic of medical domain.
At present, used means mainly are divided into two kinds in the blood disease diagnosis: a kind of is to be used widely over past ten years, according to the automatic hematology analyzer and the method (AHA) of all principle design manufacturings of resistance variations, laser light scattering, radio frequency, histochemistry and flow cytometer.This method is simple, quick and can carry out multiparameter mensuration, but this quasi-instrument is mainly classified according to the volume size of haemocyte, concrete form, design feature to various cells can not be discerned, and available information is less, generally can only realize three classification or five classification.Say exactly cry and hive off, and do not cry classification.The many of medical system studies show that this quasi-instrument classification is powerless to the abnormal cell of leukaemia and various anemias, has covered many effective medical diagnosis on disease information.Even the automatic hematology analyzer of higher class and method also only can be as the screenings of normal specimen.If do not match, be easy to produce and fail to pinpoint a disease in diagnosis or mistaken diagnosis with smear for microscopic examination.Another kind is to dye by marrow cell coating plate, then at microscopically hand control manual machinery shifter, the searching blood cell detects by an unaided eye, kind and number thereof to haemocyte are observed, are judged, thisly check that one by one the inspection work of cell is very loaded down with trivial details and tired, need a lot of manpowers, and need special medical personnel to discern; Because some cellular change difference is very little, and lacks objective decision metrics, subjective error can't be avoided, and also can't form effective quantitative description result simultaneously.In today that mass screening is progressively carried out, traditional inspection method is incompatible to be needed, and people need automatically to finish the instrument of these work, promptly realize the method for the automatic identification of blood cell with computing machine.
The nearest development of this quasi-instrument is that the instrument of first kind of mode comprehensive two kinds and even multiple signal improve classification quality in the world, as automatic leukocyte analysis instrument of the VCS of Coulter company and method synthesis measure analysis, high frequency conduction analyze and laser light scattering is analyzed three kinds of technology.Studies show that it helps antidiastole part blood disease really, but can not make the marrow blood cytological classification.The development of the second way is to have adopted automatic image analyzer and method device, by the form and the parametric texture of computing machine extraction haemocyte and karyon, utilizes artificial intelligence imitation human eye pair cell to classify automatically.State such as the U.S. and Israel has all released multiple commercial equipment, and price is about 400,000 dollars.Domestic from last century the eighties take to the research of cyto-diagnosis aspect, set up the automatic Study of recognition cooperative groups of cancer cell, obtained some initial achievements.The early 1990s, the Luo Li people of Southeast China University etc. utilize microcomputer and monochromatic area array CCD to gather the leucocyte image, than systematic research the leucocyte automatic recognition problem, but owing to adopt monochromatic gray level image, make the cell segmentation character description method have certain difficulty, cause recognition accuracy not ideal, thereafter, development along with computer technology and colored area array CCD imaging technique, Wang Dinghui, people such as Zhang Yong are applied to the blood cell automated Classification with Color Image Processing and analytical technology, make system recognition rate be improved, still present these achievements in research also fail to obtain practical application.
First kind of mode utilized the physical chemistry of cell to form information and cell size information, but can't utilize the abundant shape information of cell as the blood disease doctor; Second kind of form and texture information that utilizes cell but can't utilize the composition change information of cell itself.And first kind of mode only be applicable to invention Peripheral blood examination, is not suitable for the marrow blood cell.Therefore all can't reach replacement artificial bone-marrow (blood) cytological classification at present, can only be as a kind of means of sieving.Use this quasi-instrument, but patient's smear reinspection rate is between 30-50%.
This situation has fundamentally been broken in the appearance of light spectrum image-forming technology for the first time.It makes us can obtain the spectrogram of marrow (blood) cell again when obtaining marrow (blood) cellular morphology, we can be put together the form of marrow (blood) cell and composition information simultaneously analyze.The light spectrum image-forming technology is applied to remote sensing analysis (claiming multispectral imaging or the imaging of super spectrum again) the earliest, and it makes us can identify forest, wheatland, meadow according to spectral information when obtaining ground photograph, and even is the mineral reserve below the face of land.The device that is used for the microspectrum imaging moved to maturity later on gradually from 98 years.External numerous research worker is progressively improving the performance of device and is enlarging application.
Multi-optical spectrum imaging technology on the biomedicine can be used to strengthen the information of obtaining, and strengthens the recognition capability of computing machine pair cell.In principle, when Normocellular character changes, some materials in its nucleus, content as DNA (deoxyribonucleic acid) (DNA) also becomes thereupon, cause nuclear form, size and caryoplasm ratio to change, and the type that changes is according to the kind of cell and grade malignancy and different; Thereby, can only obtain the identifying information of cell 60%-70% at most to general cell image gray scale and morphologic analysis; Can not show a candle to multispectral image in the observation of cell form, by the imaging spectral analysis, obtain forming in the cell information (it is directly relevant with light absorbing material composition to be absorbed light wavelength) that changes, these cytochemistry component quantifying analytical informations make the original computer image analysis method that very application prospects arranged.More domestic scholars have studied blood of human body leucocyte spectrum with fiber spectrometer on microscope, they have noticed that patient's lymphocyte has visibly different spectral characteristic with granulocyte with comparing of normal person, and same a kind of spectrum from cell of dissimilar patients is also inequality.This shows, utilize microspectrum to carry out spectrum identification and classification, can be leukemia diagnosis new spectroscopy foundation is provided leucocyte.
Summary of the invention
The present invention seeks to: multi-optical spectrum imaging technology is combined with microtechnic, utilize novel electric tuning LCD filters (LCTF) to replace traditional light filter to realize the quick scanning of wavelength, adopt high sensitivity refrigeration area array CCD that marrow (blood) cell Wright's staining is coated with picture and carry out digitizing.It has obtained the spectrogram of marrow (blood) cell again when obtaining marrow (blood) cellular morphology.Utilize feature (the homocellular general character of different haemocytes in research normal person marrow (blood) smear of light spectrum image-forming technological system, the otherness of different cells, and extraneous factor such as sample preparation, light source is to the influence and the removing method of spectrum), the variation of the spectral characteristic of cell after the generation pathology, the morphological parameters characteristics of haemocyte etc. are for the research of hemopathic early diagnosis and automated diagnostic is laid a solid foundation.By the imaging spectral analysis, can obtain in the cell forming the information that changes, thereby make more accurately fast the analysis identification of not sick cell and sick cell.With only utilize the blood automatic image analyzer of cellular morphology to compare at present in the world with method, this automation equipment degree and precision of analysis and reliability all are greatly improved.Compare with hand inspection, except alleviating complicated hand labour, detect accuracy rate and will improve greatly, will make more patient just can find disease in early days, reach the purpose of healing, its social benefit is huge.Order only utilizes the blood automatic image analyzer of cellular morphology to compare with method in the world, detects accuracy rate and is significantly improved.
A kind of micro-multispectral marrow involved in the present invention and the technical purpose of peripheral blood cells automatic analyzer and method be achieved in that it comprise light source, microscope, display and, area array CCD and computing machine; It is characterized in that; A, a filter system is arranged, its core component is the electric tuning light filter, it and electric tuning control device of the light filter and area array CCD are configured on the microscope, be used to realize the quick collection of microspectrum image, computing machine is then regulated the voltage that is added on the electric tuning light filter by control electric tuning control device of the light filter, changes the centre frequency of its passband; B, CCD control and data acquisition system (DAS) are arranged, be used for realizing that converting the detected picture signal of area array CCD to digital picture by image pick-up card is stored in computing machine, carries out Flame Image Process and analysis for computing machine; C, a three-dimensional mobile control system is arranged, be used for realizing automatic focus and autoscan under computer control, automatic carrier is connected with computing machine by three-dimensional mobile control system, is passed through to scan by computing machine; D, also have computing machine, be used for to the control of three-dimensional mobile control system and to the spectrum picture that collects carry out that image is synthetic, image rectification, cut apart automatically, Classification and Identification automatically.The image calculation of area array CCD collection focuses on, the control automatic carrier moves up and down and carries out automatic focus, moves horizontally the realization smear
Technique effect of the present invention: compared with prior art, have outstanding characteristics and obvious improvement.Wherein, with only utilize the blood automatic image analyzer of cellular morphology to compare at present in the world with method, automaticity, analysis result parasexuality, reliability increase significantly, make full use of microspectrum haemocyte is carried out spectrum identification and classification, for hemopathic diagnosis provides new spectroscopy foundation; Compare with hand inspection, except alleviating complicated hand labour, detect accuracy rate and improve greatly.The present invention makes more patient just can find disease in early days, reaches the purpose of healing, and its social benefit is huge.
Description of drawings
Fig. 1 is the structure and the method synoptic diagram of whole instrument, wherein: 1-microscope 2-light source, 3-automatic carrier, the three-dimensional mobile control system 5-of 4-electric tuning control device of the light filter, 6-printer and other peripheral hardware, the 7-computing machine, 8-data backup restoration system, 9-display, the 10-picture monitor, 11-CCD control and data acquisition system (DAS), 12-area array CCD, 13, the electric tuning light filter.
Fig. 2 is system works flow process figure
Fig. 3 is the multi-optical spectrum image collecting process flow diagram
Fig. 4 is the typical marrow image of a width of cloth,
Fig. 5 is among Fig. 4 in the rectangular area at the bottom of average back of the body, ripe red blood cell, cytoplasm and the nuclear curve of spectrum
Fig. 6 is automatic partitioning algorithm process flow diagram
Embodiment
The present invention will be further described below in conjunction with accompanying drawing:
At Fig. 1 is an embodiment of automatic analyzer of the present invention and method.It has a light source 2, microscope 1, display 9 and area array CCD 12 and computing machine 7; Also have: a, a filter system is arranged, its core component is an electric tuning light filter 13, it and electric tuning control device of the light filter 5 and area array CCD 12 are configured on the microscope 1, be used to realize the quick collection of microspectrum image, 7 in computing machine is regulated the voltage that is added on the electric tuning light filter 13 by control electric tuning control device of the light filter 5, changes the centre frequency of its passband; B, CCD control and data acquisition system (DAS) 11 are arranged, be used for realizing that converting area array CCD 12 detected picture signals to digital picture by image pick-up card is stored in computing machine 7, carries out Flame Image Process and analysis for computing machine; C, a three-dimensional mobile control system 4 is arranged, be used under computer control, realizing automatic focus and autoscan, automatic carrier 3 is connected with computing machine 7 by three-dimensional mobile control system 4, and the image calculation of being gathered by area array CCD 12 by computing machine 7 focuses on, control automatic carrier 3 moves up and down and carries out automatic focus and move horizontally realizing smear scanning; D, also have computing machine 7, be used for to the control of three-dimensional mobile control system (4) and to the spectrum picture that collects carry out that image is synthetic, image rectification, cut apart automatically, Classification and Identification automatically; Say further: described electric tuning light filter 13 1 ends are installed on the microscope 1 by the standard C interface, and the other end is connected with area array CCD 12, are imaged on the area array CCD 12 behind the stain smear cell picture process electric tuning light filter 13; Electric tuning device 13 is mounted in the place ahead of area array CCD 12, is used to compensate the optical path length lens barrel by one and is installed in microscope 1; After the object picture amplifies by microscope, again by 13 imaging of electric tuning light filter on area array CCD 12; Control and the data acquisition system (DAS) 11 of area array CCD 12 detected picture signals by area array CCD converts digital picture to and is stored in the computing machine 7, carries out Flame Image Process and analysis for computing machine 7; Electric tuning light filter 13 is as filtering device, its bandwidth is 5-30nm, can be the liquid crystal tunable filter, also can be the electric tuning light filter of other types, be added in voltage on the electric tuning light filter 13 by computer control, change the optical wavelength that sees through electric tuning light filter 13; Smear under the irradiation of continuous light, the light signal that cell or other small items produced (transmitted light) by electric tuning light filter 13 after, be imaged on the area array CCD 12; Under the control of computing machine 7, under each visual field, can be on several wave bands of choosing each wave band gather piece image, then by 7 pairs of spectrum pictures that collected of computing machine carry out that image is synthetic, image rectification, cut apart automatically, operation such as Classification and Identification automatically; Light source 2 adopts kohler's illumination, guarantees the homogeneity of visual field; Under different enlargement factors, adopt different 12 time shutter of area array CCD and the different intensities of light source, under identical enlargement factor, adopt identical 12 time shutter of area array CCD and the identical intensity of light source, guaranteed the consistance of acquisition condition.
Fig. 2 is system works flow process figure.
Open the light source 2 and the electric tuning control device of the light filter 5 of microscope 1, treat that approximately light source 2 is stable half an hour after, just can start working.At first gather background spectrum, the multi-optical spectrum image collecting method that method is seen below.After marrow (blood) smear was put on the automatic carrier 3, Manual focusing was transferred to correct position; Start computing machine 7 automatic scanners; Under the control of computing machine 7, be automatically moved to a visual field, the transmission peak wavelength of electric tuning light filter 13 is navigated to the middle wavelength of the wavelength that will gather, under this wavelength, gather piece image, the image that collects is analyzed, utilize image process method to judge whether slide focuses on, if do not focus on, so at the height of constantly regulating automatic carrier 3 under the control of computing machine 7 till focusing on; After a visual field focuses on, just can gather a multispectral image, after the multi-optical spectrum image collecting of a visual field finished, computing machine 7 carried out image segmentation automatically, finds out cytoplasm and nucleus; Image is after segmenting, can carry out cell characteristic measures, extract some morphological features and spectral signature, carry out cytological classification with these features, after visual field is cut apart and analysis finishes, continue to gather and analyze next visual field again, after visual field number to be collected reaches certain requirement, stop images acquired, the cell of all collections of front is carried out analysis-by-synthesis, utilize neural network classifier to draw classification of diseases according to the disease database of under physician guidance, setting up.
Fig. 3 is the multi-optical spectrum image collecting process flow diagram.Owing to spectrum picture is combined by a series of monochrome image, therefore, need under each our wavelength selected, gather piece image under each visual field; Open light source 2 and electric tuning control device of the light filter 5, behind light stability half an hour, marrow (blood) smear is placed on the automatic carrier 3 approximately, light source 2 is adjusted to a suitable position, later all be put into identical position for identical enlargement ratio at every turn, make illumination intensity identical; The light source of regulating microscope 1 is to kohler's illumination, and is even with the brightness that guarantees the visual field; The method of image acquisition is under each visual field, at first the transmission peak wavelength of electric tuning light filter 13 is navigated to the middle wavelength of the wavelength that we will gather, under the control of computing machine 7, utilize image processing algorithm to carry out automatic focus, make the visual field the most clear, switch under the selected wave band by computing machine 7 control electric tuning light filters 5 then, gather piece image, the image that collects is carried out gray correction, and then switch to next selected wave band and gather piece image again and carry out gray correction again, so circulation, finish up to whole collection of all selected wave bands, carry out the synthetic and data storage of colour again, utilize computing machine 7 image processing techniquess to find out the karyon and the endochylema of haemocyte automatically, measure the characteristic parameter of cell, utilize neural network method to draw the affiliated classification of cell, deposit database in the lump in together with characteristic parameter.Gather next visual field again, reach requirement up to the scanning field of view number; When all interesting areas of a smear all gather finish after, just can carry out automatical analysis to the data of measuring, whether provide information such as pathology information, number percent, optimum grade malignancy;
Because the blue light that incandescent source is rendered as in the spectrum in spectrum output is lower, and near infrared (NIR) is higher; The quantum efficiency of area array CCD (QE) curve normally at blue region seldom, and is and the highest at ruddiness.The flux characteristic of electric tuning light filter 13 (throughputproperties) also shows as certain curve of spectrum.For example, the LCTF light filter sees through less at blue light, and high in NIR part transmitance.Therefore, we can meet the trouble at the combined influence of the low performance of blue light.This problem can be eased by being chosen in the area array CCD that blue portion has higher QE, for example a thin back-illuminated type (backilluminate) area array CCD.But just need a high performance camera in this case, this will cause the raising of price.The method that the present invention proofreaies and correct by software has well solved the problem of incandescent source, also effectively eliminates the influence to imaging spectral of excitation source 2, microscope 1, electric tuning light filter 13, detecting device, guarantees the consistance that imaging detects;
The method of gray correction is; Focus on smear, make cell the best clear; Mobile smear is to white space, under an empty visual field, defocus slightly, by computing machine 7 control electric tuning light filter 5 and area array CCDs 12, with gathering piece image at each selected wave band an identical integral time, calculate each wave band next suitable integral time according to these images again, can give a long integral time for the wave band that gray-scale value is lower, give integral time of a weak point for the high wave band of gray-scale value, to carry out balanced compensated to whole wavelength band.According under each selected wave band, gathering piece image again, spectrum picture at the bottom of the back of the body that is enhanced again the integral time that calculates.Later image for each wave band, all use gather the integral time identical with spectrum picture at the bottom of the back of the body that strengthens, all corresponding with spectrum picture at the bottom of the back of the body that strengthens band image carries out division operation and then multiply by a scale factor gray-scale value is promoted, can guarantee after handling like this that all there is consistent gray-scale value in the back bottom area territory at different wave bands, is convenient to Flame Image Process and analysis.Because the correction to the back of the body end is that unit handles with the pixel, so not only can effectively remove the imaging system influence different to different-waveband sensitivity, also can compensate lifting by a pair of blue wave band, effectively eliminate some slight visual field unevenness and area array CCD 12 each pixel luminous sensitivity difference.
Fig. 4 is the typical marrow image of a width of cloth, and Fig. 5 is among Fig. 4 at the bottom of the back of the body in the rectangular area, ripe red blood cell, cytoplasm and the nuclear average curve of spectrum.The horizontal ordinate of Fig. 5 is a wavelength, and ordinate is the transmission light intensity, gathers piece image every 10nm, the average gray in the rectangular area of the monochrome image in each some representative graph 4 in each wave band.
As can be seen from Figure 5, different materials, their spectral-transmission favtor is different.We can utilize these differences that they are cut apart automatically.Because nucleus, cytoplasm have different spectral characteristics with the background area, utilize their SPECTRAL DIVERSITY, adopt and image is cut apart based on the image partition method of pixel, obtained good segmentation effect.
The main dividing method spectral ratio operation of adopting.Because the color distinction that dissimilar bone marrow cells have is less, and, utilizes the difference of some wave bands merely or utilize the absolute intensity value of image also to be difficult to distinguish accurately at the bottom of nucleus, cytoplasm and the back of the body because the dyeing depth is also very difficult in full accord.Therefore, we adopt the method for spectrum ratio to strengthen difference between them, and ratio is relative value, and this has just reduced the dyeing depth and the photosensitive unit sensitivity influence to image segmentation result greatly.Concrete grammar is to select two wave bands, to each image pixel, utilize the gray scale of these two wave bands to carry out the phase division operation, again the result who obtains be multiply by a fixing amplification factor, generate the new monochrome image of a width of cloth, utilize this monochrome image to carry out automatic threshold again and cut apart, what the present invention used is the maximum variance threshold method, has obtained good effect.The present invention has found several groups of wavelength, can be partitioned into cytoplasm and nucleus respectively;
Because the mass data of spectrum picture, band selection are to handle a necessary link of spectrum picture, band selection can effectively reduce calculated amount on the one hand, has also removed redundant information wherein on the other hand.The searching algorithm of feature selecting can be divided into optimum search and suboptimum is searched for two classes, and optimal algorithm has the method for exhaustion and branch-bound algorithm; Sub-optimal algorithm then comprises simulated annealing, Tabu algorithm and genetic algorithm or the like.From can obtaining optimal result and realize simple angle, the present invention adopts the method for exhaustion.It is to travel through all possible set of wavelengths with the method for exhaustion that the present invention looks for the method for wavelength, corresponding different set of wavelengths, obtain the ratio of the gray scale of these two kinds of material correspondences respectively, the constitutive characteristic space, the separability criterion of obtaining this two classes material then is (as between class distance in the class, Bhattacharyya distance and divergence), the error rate during match stop, one group of wavelength of error rate minimum is institute and asks.
Fig. 6 is that marrow (blood) smear cell image is cut apart (detection) process flow diagram automatically
The present invention has studied marrow (blood) smear cell image cuts apart (detection) algorithm automatically, has found a series of full-automatic Fast Segmentation Algorithm.Image segmentation mainly adopts the method for spectrum ratio.Because nucleus, cytoplasm, the ripe red blood cell and the back of the body end, have different spectral characteristics, therefore, can utilize their spectral characteristic to carry out full-automatic image segmentation.Specifically cutting apart flow process is: choose several wave bands, utilize these band class information to find out white space; Select some wave bands again, utilize band class information to find out ripe red blood cell zone; Just can obtain the karyocyte zone behind deletion red blood cell zone and the blank back bottom area territory.Carry out two-value morphological operations such as corroding expansion again and remove some little particles, obtain region S 1.Select two wave bands, carry out the spectral ratio operation, the ratio image that obtains, utilize the maximum variance threshold method to find the threshold value of an overall situation in the zone of S1 institute mark, can separate endochylema and karyon, obtain karyon region S 2, the effect of cutting apart like this is more rough, but speed is very fast.Deletion S2 medium and small surface area, the part of deletion is filled with the endochylema mark, and then the zone that deletion does not link to each other with karyon obtains region S 3, and S3 has been exactly a cell one by one so.Because be rough cutting apart cutting apart of front, therefore, selects several wave bands again, as cutting apart unit, utilizes band class information that it is accurately cut apart with each cell, can be partitioned into the karyon and the endochylema of each cell accurately.After cell segmentation is intact, just can carry out Classification and Identification.
The effect of utilizing visual method to come picture with the aid of pictures to cut apart.For peripheral blood film, about 8000 cells to Wright dyeing are cut apart, the accuracy that the nucleus of monocyte, lymphocyte, band form neutrophilic granulocyte, neutrophilic segmented granulocyte, eosinophil is cut apart is more than 98.7%, and the accuracy that the basophilic granulocyte nucleus is cut apart is 90.2%.The accuracy that its cytoplasm is cut apart is a little bit poorer slightly, mainly is that its SPECTRAL DIVERSITY is minimum because the cytoplasm that has is compared with other akaryote, has also reached more than 90% but cut apart accuracy rate, far above having the value that document is reported now.Marrow cell coating plate for Wright dyeing, about 9000 cells are cut apart, common metarubricyte, rubricyte, early erythroblast, neutral leaflet nuclear, neutral band form nucleus, neutrophilic myelocyte, neutrophilic metamyelocyte, the accuracy that its nucleus of lymphocyte is cut apart is more than 98.5%, for fewer acidophilia and basicyte, the accuracy that nucleus is cut apart is more than 95.1%, for cutting apart of endochylema, except late children is red, accuracy rate is all more than 94.3%, because the late young red ripe red blood cell that just become when growing up, so the spectral characteristic of its endochylema is very approaching with ripe red blood cell, so occur wrong situation of dividing easily, from present case, accuracy rate has also reached 85.7%.
The final purpose of image segmentation is in order to carry out Classification and Identification.In order can effectively to discern, must set up feature database earlier.The method of setting up feature database is, collect case slide 300 examples of normal population, each phase type of leukemia earlier, gather their micro-multispectral image, and under the expert instructs, carry out the manual sort, program and manually accurately cut apart and measure, carry out feature extraction.Because the parameter of extracting is numerous, and multispectral image has a lot of wave bands, and therefore, its spectral signature has more several times than common gray level image.And too many feature not only can make the speed of analysis slack-off, and may make the result of analysis more error occur.Therefore, carry out the screening of feature selecting and feature, the dimension that reduces feature is necessary.For training sample, the present invention has extracted morphology parameter, optical density parameter and parametric texture and the spectral ratio parameter amounts to 438 features, the present invention adopts genetic algorithm to carry out feature selecting, analyze the spectrum and the morphological character of dissimilar leukaemia, dissimilar cells, analyze their specificity and correlativity, select the characteristic parameter of cytological classification, obtained good effect.Through genetic algorithm, obtain 132 of optimal characteristics at last, after feature selecting finished, we set up the initial characteristic data storehouse with regard to utilizing the feature after selecting.
After the primitive character storehouse has been set up, make up 3 layers of neural network, utilize neural network, after network training finishes the weights data of training are deposited in database these sample training.
The result of primitive character storehouse and neural metwork training has been arranged, and we just can discern unknown sample.The present invention adopts neural network that cell to be measured is carried out automatic recognition classification, obtained good result, and speed is exceedingly fast.From recognition effect, for peripheral blood film, about 8000 cells to Wright dyeing are classified, the monocyte accuracy is 97.2%, the lymphocyte accuracy is 98.7%, the band form neutrophilic granulocyte accuracy is 95.3%, the neutrophilic segmented granulocyte accuracy is 96.6%, the eosinophil accuracy is 97.6%, and the basophilic granulocyte accuracy is 92.1%.Marrow cell coating plate for Wright dyeing, about 9000 cells are classified, accuracy rate is: common metarubricyte 94.6%, rubricyte 93.2%, early erythroblast 90.7%, neutral leaflet nuclear 95.5%, neutral band form nucleus 87.4%, neutrophilic myelocyte 89.4%, neutrophilic metamyelocyte 88.3%, lymphocyte 92.0%, for fewer acidophilia and basicyte and other cell all more than 80.3%.
Claims (10)
1, a kind of micro-multispectral marrow and peripheral blood cells automatic analyzer and method, it comprises light source (2), microscope (1), display (9) and area array CCD (12) and computing machine (7); It is characterized in that: a, a filter system is arranged, its core component is electric tuning light filter (13), it and electric tuning control device of the light filter (5) and area array CCD (12) are configured on the microscope (1), be used to realize the quick collection of microspectrum image, computing machine (7) is then regulated the voltage that is added on the electric tuning light filter (13) by control electric tuning control device of the light filter (5), changes the centre frequency of its passband; B, CCD control and data acquisition system (DAS) (11) are arranged, be used for realizing that converting the detected picture signal of area array CCD (12) to digital picture by image pick-up card is stored in computing machine (7), carries out Flame Image Process and analysis for computing machine; C, a three-dimensional mobile control system (4) is arranged, be used under computer control, realizing automatic focus and scanning, automatic carrier (3) is connected with computing machine (7) by three-dimensional mobile control system (4), and the image calculation of being gathered by area array CCD (12) by computing machine (7) focuses on, control automatic carrier (3) moves up and down and carries out automatic focus and move horizontally realizing smear scanning; D, also have computing machine (7), be used for to the control of three-dimensional mobile control system (4) and to the spectrum picture that collects carry out that image is synthetic, image rectification, cut apart automatically, Classification and Identification automatically.
2, by described analyser of claim 1 and method, it is characterized in that, described electric tuning light filter (13) one ends are installed on the microscope (1) by the standard C interface, the other end is connected with area array CCD (12), is imaged on the area array CCD (12) behind the stain smear cell picture process electric tuning light filter (13).
3, by described analyser of claim 1 and method, it is characterized in that the bandwidth of electric tuning light filter (13) is 5-30nm, can be the liquid crystal tunable filter.
4, by described analyser of claim 1 and method, it is characterized in that, light source (2) adopts kohler's illumination, be used to guarantee the homogeneity of visual field, it is to adopt different area array CCD (12) time shutter and the different intensities of light source under different enlargement factors, under identical enlargement factor, adopt identical area array CCD (12) time shutter and the identical intensity of light source, guarantee the consistance of acquisition condition.
5, by described analyser of claim 1 and method, it is characterized in that its working routine comprises the steps: a, automatic focus; The collection of b, spectrum picture; C, based on the cell segmentation of spectrum picture; D, characteristics of image are measured; E, cytological classification; F, autoscan; G, classification of diseases.
By described analyser of claim 5 and method, it is characterized in that 6, the collection of described spectrum picture and background correction comprise the steps: a, gather a width of cloth background image at each selected wave band earlier; B, calculate integral time of each spectral band according to the background image that is collected; C, again according to integral time of calculating gathering the background image that a width of cloth strengthens at each selected wave band; D, the image that collects later on use identical with the background image that strengthens integral time at each wave band, the image of each wave band all looks like to carry out gray correction according to the back of the body base map that strengthens, to eliminate or to reduce the influence of light source, optical element, area array CCD (12) photosensitive unit, luminous sensitivity and spectral response.
7, by described analyser of claim 5 and method, it is characterized in that, described cell segmentation detects, adopt a kind of full automatic spectrum dividing method, it is to adopt the method for spectral ratio to obtain some gray level images, utilizes automatic threshold dividing method and two-value morphological method to carry out by thick to smart cutting apart again.
8, by described analyser of claim 7 and method, it is characterized in that, described full automatic spectrum is cut apart flow process and is: choose several wave bands, utilize these band class information to find out white space, select some wave bands again, utilize band class information to find out ripe red blood cell zone, just can obtain the karyocyte zone behind deletion red blood cell zone and the blank back bottom area territory; Carry out two-value morphological operations such as corroding expansion again and remove some little particles, obtain region S 1; Select two wave bands, carry out the spectral ratio operation, the ratio image that obtains, utilize the maximum variance threshold method to find the threshold value of an overall situation in the zone of S1 institute mark, can separate endochylema and karyon, obtain karyon region S 2, the effect of cutting apart like this is more rough, but speed is very fast; Deletion S2 medium and small surface area, the part of deletion is filled with the endochylema mark, and then the zone that deletion does not link to each other with karyon obtains region S 3, and S3 is exactly a cell one by one so; Because be rough cutting apart cutting apart of front, therefore, selects several wave bands again, as cutting apart unit, utilizes band class information that it is accurately cut apart with each cell, can be partitioned into the karyon and the endochylema of each cell accurately.After cell segmentation is intact, just can carry out Classification and Identification.
9, by described analyser of claim 6 and method, it is characterized in that, the band selection of described micro-multispectral analysis, it is to utilize the separability criterion, selects one group of wave band can distinguishing two kinds of materials with the method for exhaustion from the wave band of setting.
10, by described analyser of claim 5 and method, it is characterized in that, described classification of diseases: a, set up the initial characteristic data storehouse: it is according to collecting normal population earlier, the case slide of each phase type of leukemia, gather micro-multispectral image, carry out the manual sort, program and manually accurately cut apart and measure, carry out feature extraction, screening, extract the morphology parameter, optical density parameter and parametric texture and spectral ratio parameter attribute, analyze dissimilar leukaemia, the spectrum of dissimilar cells and morphological character, analyze their specificity and correlativity, select the characteristic parameter of cytological classification, obtain optimal characteristics at last, the feature after will selecting is again set up the initial characteristic data storehouse; B, 3 layers of neural network of structure utilize neural network to these sample training, after network training finishes the weights data of training are deposited in database; C, according to the result of primitive character storehouse and neural metwork training, unknown sample is discerned, with neural network cell to be measured is carried out automatic recognition classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101116115A CN1296699C (en) | 2003-12-19 | 2003-12-19 | Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101116115A CN1296699C (en) | 2003-12-19 | 2003-12-19 | Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1553166A true CN1553166A (en) | 2004-12-08 |
| CN1296699C CN1296699C (en) | 2007-01-24 |
Family
ID=34336221
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2003101116115A Expired - Fee Related CN1296699C (en) | 2003-12-19 | 2003-12-19 | Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1296699C (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101126699B (en) * | 2007-09-27 | 2010-07-21 | 中国气象局北京城市气象研究所 | Pollen grain image data collecting system |
| CN101907559A (en) * | 2009-06-02 | 2010-12-08 | 蓝伊精机株式会社 | Specimen processing apparatus and specimen processing method |
| CN101339185B (en) * | 2008-06-23 | 2011-03-02 | 武汉呵尔医疗科技发展有限公司 | Automatic microscopic imager for detecting cast-off cells and detection method |
| CN102359938A (en) * | 2011-09-16 | 2012-02-22 | 长沙高新技术产业开发区爱威科技实业有限公司 | Morphological analytical apparatus and method for erythrocytes |
| CN102625022A (en) * | 2011-01-29 | 2012-08-01 | 麦克奥迪实业集团有限公司 | Rapid microscopic section scanning method with image space compensation motion and apparatus thereof |
| CN102822671A (en) * | 2010-03-30 | 2012-12-12 | 日本电气株式会社 | Image processing apparatus, image reading apparatus, image processing method and image processing program |
| CN103146800A (en) * | 2013-03-27 | 2013-06-12 | 朱耀辉 | Full-automatic cell non-staining image recognizing and counting method |
| CN105027165A (en) * | 2013-03-15 | 2015-11-04 | 文塔纳医疗系统公司 | Tissue object-based machine learning system for automated scoring of digital whole slides |
| CN105319381A (en) * | 2015-11-16 | 2016-02-10 | 青岛安倍康生物医药技术有限公司 | Full-automatic blood cell counter and counting control method thereof |
| US9367913B2 (en) | 2010-09-17 | 2016-06-14 | Koninklijke Philips N.V. | Choosing anatomical variant model for image segmentation |
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN108140239A (en) * | 2015-09-23 | 2018-06-08 | 皇家飞利浦有限公司 | For organizing the method and apparatus of identification |
| CN108693096A (en) * | 2017-12-13 | 2018-10-23 | 青岛汉朗智能医疗科技有限公司 | Red blood cell detection method and system |
| CN108885682A (en) * | 2016-02-26 | 2018-11-23 | 谷歌有限责任公司 | Use Processing with Neural Network cell image |
| CN108896363A (en) * | 2018-05-28 | 2018-11-27 | 杭州智微信息科技有限公司 | A kind of digitized flow and method of bone marrow smear |
| CN109211803A (en) * | 2018-09-17 | 2019-01-15 | 中国科学院生态环境研究中心 | A kind of device that micro- plastics are quickly identified based on micro- multispectral technology |
| CN109324197A (en) * | 2018-11-16 | 2019-02-12 | 北京诚名扬科技发展有限公司 | Bracket for blood grouping recorder and identification recording method |
| CN109473151A (en) * | 2018-10-31 | 2019-03-15 | 青岛大学 | A kind of hospital clinical inspection blood collection storage device and its acquisition method |
| CN109489816A (en) * | 2018-10-23 | 2019-03-19 | 华东师范大学 | A kind of micro- high light spectrum image-forming platform and the method for big area data cube acquisition |
| CN109612997A (en) * | 2019-01-22 | 2019-04-12 | 董玮 | A kind of pathology department's clinic etiologic diagnosis device |
| CN110411717A (en) * | 2019-08-30 | 2019-11-05 | 广东北创光电科技股份有限公司 | Micro- spherical surface spectrum automatic analyzer and the spectrum automatic analysis method for applying it |
| CN110767307A (en) * | 2019-09-20 | 2020-02-07 | 杭州憶盛医疗科技有限公司 | Novel artificial intelligent automatic diagnosis method for cell morphology |
| CN111458269A (en) * | 2020-05-07 | 2020-07-28 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligent identification method for peripheral blood lymph micronucleus cell image |
| CN111504885A (en) * | 2020-04-04 | 2020-08-07 | 电子科技大学 | Full-automatic blood smear morphological analysis device based on machine vision |
| 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 |
| CN111795918A (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 scanning structure and scanning method |
| CN112070059A (en) * | 2020-09-18 | 2020-12-11 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligent classification and identification method for blood cell and marrow cell images |
| CN112378727A (en) * | 2020-12-14 | 2021-02-19 | 湖南莱博赛医用机器人有限公司 | DNA ploid quantitative analysis device based on machine vision and application method thereof |
| CN112396583A (en) * | 2020-11-18 | 2021-02-23 | 深思考人工智能机器人科技（北京）有限公司 | Method and system for DNA (deoxyribonucleic acid) ploid quantitative analysis based on Papanicolaou staining mode |
| CN112432948A (en) * | 2020-12-03 | 2021-03-02 | 中国人民解放军陆军军医大学第二附属医院 | Automatic detection method for peripheral blood cell morphology |
| CN113039551A (en) * | 2018-11-19 | 2021-06-25 | 深圳迈瑞生物医疗电子股份有限公司 | Method of analyzing cells, cell analysis apparatus, and computer-readable storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2032446U (en) * | 1988-01-12 | 1989-02-15 | 华东师范大学 | Intelligent corpuscle detector |
| US4965725B1 (en) * | 1988-04-08 | 1996-05-07 | Neuromedical Systems Inc | Neural network based automated cytological specimen classification system and method |
| JPH09197290A (en) * | 1996-01-12 | 1997-07-31 | Omron Corp | Microscope image input device |
| CN1277394A (en) * | 2000-07-05 | 2000-12-20 | 徐秀林 | Recognition system and analysis method for microscopic biological and medical images |
| CN1148574C (en) * | 2001-03-08 | 2004-05-05 | 武汉大学 | High resolution Adama conversion micro-image analyzer |
-
2003
- 2003-12-19 CN CNB2003101116115A patent/CN1296699C/en not_active Expired - Fee Related
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101126699B (en) * | 2007-09-27 | 2010-07-21 | 中国气象局北京城市气象研究所 | Pollen grain image data collecting system |
| CN101339185B (en) * | 2008-06-23 | 2011-03-02 | 武汉呵尔医疗科技发展有限公司 | Automatic microscopic imager for detecting cast-off cells and detection method |
| CN101907559A (en) * | 2009-06-02 | 2010-12-08 | 蓝伊精机株式会社 | Specimen processing apparatus and specimen processing method |
| CN102822671B (en) * | 2010-03-30 | 2014-07-02 | 日本电气株式会社 | Image processing apparatus, image reading apparatus, image processing method and image processing program |
| US9438768B2 (en) | 2010-03-30 | 2016-09-06 | Nec Corporation | Image processing apparatus, image reading apparatus, image processing method and information storage medium |
| CN102822671A (en) * | 2010-03-30 | 2012-12-12 | 日本电气株式会社 | Image processing apparatus, image reading apparatus, image processing method and image processing program |
| US9367913B2 (en) | 2010-09-17 | 2016-06-14 | Koninklijke Philips N.V. | Choosing anatomical variant model for image segmentation |
| CN102625022B (en) * | 2011-01-29 | 2014-05-28 | 麦克奥迪实业集团有限公司 | Rapid microscopic section scanning method with image space compensation motion and apparatus thereof |
| CN102625022A (en) * | 2011-01-29 | 2012-08-01 | 麦克奥迪实业集团有限公司 | Rapid microscopic section scanning method with image space compensation motion and apparatus thereof |
| CN102359938B (en) * | 2011-09-16 | 2012-09-05 | 长沙高新技术产业开发区爱威科技实业有限公司 | Morphological analytical apparatus and method for erythrocytes |
| CN102359938A (en) * | 2011-09-16 | 2012-02-22 | 长沙高新技术产业开发区爱威科技实业有限公司 | Morphological analytical apparatus and method for erythrocytes |
| CN105027165A (en) * | 2013-03-15 | 2015-11-04 | 文塔纳医疗系统公司 | Tissue object-based machine learning system for automated scoring of digital whole slides |
| CN105027165B (en) * | 2013-03-15 | 2021-02-19 | 文塔纳医疗系统公司 | Tissue object-based machine learning system for automated scoring of digital whole slides |
| CN103146800A (en) * | 2013-03-27 | 2013-06-12 | 朱耀辉 | Full-automatic cell non-staining image recognizing and counting method |
| CN103146800B (en) * | 2013-03-27 | 2014-08-13 | 朱耀辉 | Full-automatic cell non-staining image recognizing and counting method |
| CN108140239A (en) * | 2015-09-23 | 2018-06-08 | 皇家飞利浦有限公司 | For organizing the method and apparatus of identification |
| CN105319381A (en) * | 2015-11-16 | 2016-02-10 | 青岛安倍康生物医药技术有限公司 | Full-automatic blood cell counter and counting control method thereof |
| CN108885682A (en) * | 2016-02-26 | 2018-11-23 | 谷歌有限责任公司 | Use Processing with Neural Network cell image |
| US11915134B2 (en) | 2016-02-26 | 2024-02-27 | Google Llc | Processing cell images using neural networks |
| US11443190B2 (en) | 2016-02-26 | 2022-09-13 | Google Llc | Processing cell images using neural networks |
| CN108885682B (en) * | 2016-02-26 | 2022-03-22 | 谷歌有限责任公司 | Processing cell images using neural networks |
| CN107505698B (en) * | 2017-07-17 | 2024-01-12 | 中国家用电器研究院 | Biological identification counting device and method |
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN108693096A (en) * | 2017-12-13 | 2018-10-23 | 青岛汉朗智能医疗科技有限公司 | Red blood cell detection method and system |
| JP2021517642A (en) * | 2018-05-28 | 2021-07-26 | 杭州智微信息科技有限公司Ｈａｎｇｚｈｏｕ Ｚｈｉｗｅｉ Ｉｎｆｏｒｍａｔｉｏｎ Ｔｅｃｈｎｏｌｏｇｙ Ｃｏ．， Ｌｔｄ． | Methods for Digitizing Bone Marrow Smear Specimens |
| AU2019277871B2 (en) * | 2018-05-28 | 2021-07-08 | Hangzhou Zhiwei Information Technology Co., Ltd. | Process method for bone marrow smear digitization |
| WO2019228250A1 (en) * | 2018-05-28 | 2019-12-05 | 杭州智微信息科技有限公司 | Process method for bone marrow smear digitization |
| CN108896363A (en) * | 2018-05-28 | 2018-11-27 | 杭州智微信息科技有限公司 | A kind of digitized flow and method of bone marrow smear |
| KR102489789B1 (en) | 2018-05-28 | 2023-01-17 | 항저우 즈웨이 인포메이션 테크놀로지 캄파니 리미티드 | Digitization Flow Method of Bone Marrow Plates |
| JP7004354B2 (en) | 2018-05-28 | 2022-01-21 | 杭州智微信息科技有限公司 | Methods for digitizing bone marrow smears |
| KR20200135460A (en) * | 2018-05-28 | 2020-12-02 | 항저우 즈웨이 인포메이션 테크놀로지 캄파니 리미티드 | Digitalization flow method of bone marrow plate |
| TWI746984B (en) * | 2018-05-28 | 2021-11-21 | 大陸商杭州智微信息科技有限公司 | A process method for digitalization of bone marrow smear |
| RU2755247C1 (en) * | 2018-05-28 | 2021-09-14 | Ханчжоу Чживэй Информэйшн Текнолоджи Ко., Лтд. | Method for digitising a bone marrow punctate smear |
| CN109211803A (en) * | 2018-09-17 | 2019-01-15 | 中国科学院生态环境研究中心 | A kind of device that micro- plastics are quickly identified based on micro- multispectral technology |
| CN109211803B (en) * | 2018-09-17 | 2020-10-09 | 中国科学院生态环境研究中心 | Device for rapidly identifying micro plastic based on microscopic multispectral technology |
| CN109489816A (en) * | 2018-10-23 | 2019-03-19 | 华东师范大学 | A kind of micro- high light spectrum image-forming platform and the method for big area data cube acquisition |
| CN109473151A (en) * | 2018-10-31 | 2019-03-15 | 青岛大学 | A kind of hospital clinical inspection blood collection storage device and its acquisition method |
| CN109324197A (en) * | 2018-11-16 | 2019-02-12 | 北京诚名扬科技发展有限公司 | Bracket for blood grouping recorder and identification recording method |
| CN113039551A (en) * | 2018-11-19 | 2021-06-25 | 深圳迈瑞生物医疗电子股份有限公司 | Method of analyzing cells, cell analysis apparatus, and computer-readable storage medium |
| CN113039551B (en) * | 2018-11-19 | 2024-01-23 | 深圳迈瑞生物医疗电子股份有限公司 | Method for analyzing cells, cell analysis device, and computer-readable storage medium |
| CN109612997B (en) * | 2019-01-22 | 2021-06-29 | 济南星齐医学检验有限公司 | Clinical qualitative diagnosis device that uses of pathology department |
| CN109612997A (en) * | 2019-01-22 | 2019-04-12 | 董玮 | A kind of pathology department's clinic etiologic diagnosis device |
| CN110411717A (en) * | 2019-08-30 | 2019-11-05 | 广东北创光电科技股份有限公司 | Micro- spherical surface spectrum automatic analyzer and the spectrum automatic analysis method for applying it |
| CN110411717B (en) * | 2019-08-30 | 2021-06-18 | 广东北创光电科技股份有限公司 | Automatic spectrum analyzer for microscopic spherical surface and automatic spectrum analysis method using same |
| CN110767307A (en) * | 2019-09-20 | 2020-02-07 | 杭州憶盛医疗科技有限公司 | Novel artificial intelligent automatic diagnosis method for cell morphology |
| CN111504885A (en) * | 2020-04-04 | 2020-08-07 | 电子科技大学 | Full-automatic blood smear morphological analysis device based on machine vision |
| CN111504885B (en) * | 2020-04-04 | 2022-03-15 | 电子科技大学 | Analysis method of full-automatic blood smear morphological analysis device based on machine vision |
| CN111458269A (en) * | 2020-05-07 | 2020-07-28 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligent identification method for peripheral blood lymph micronucleus cell image |
| CN111795918A (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 scanning structure and scanning method |
| 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 |
| CN111795918B (en) * | 2020-05-25 | 2024-03-01 | 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 scanning structure and scanning method |
| CN111795919B (en) * | 2020-05-25 | 2024-03-26 | 中国人民解放军陆军军医大学第二附属医院 | Automatic bone marrow cell morphology detection system and working method thereof |
| CN112070059A (en) * | 2020-09-18 | 2020-12-11 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligent classification and identification method for blood cell and marrow cell images |
| CN112396583B (en) * | 2020-11-18 | 2024-01-26 | 深思考人工智能机器人科技（北京）有限公司 | DNA ploidy quantitative analysis method and system based on Papanicolaou staining mode |
| CN112396583A (en) * | 2020-11-18 | 2021-02-23 | 深思考人工智能机器人科技（北京）有限公司 | Method and system for DNA (deoxyribonucleic acid) ploid quantitative analysis based on Papanicolaou staining mode |
| CN112432948A (en) * | 2020-12-03 | 2021-03-02 | 中国人民解放军陆军军医大学第二附属医院 | Automatic detection method for peripheral blood cell morphology |
| CN112378727A (en) * | 2020-12-14 | 2021-02-19 | 湖南莱博赛医用机器人有限公司 | DNA ploid quantitative analysis device based on machine vision and application method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1296699C (en) | 2007-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1296699C (en) | Microscopic multispectral marrow and its peripheral blood cell auto-analyzing instrument and method | |
| JP6726704B2 (en) | Cell volume and composition measurement | |
| US7587078B2 (en) | Automated image analysis | |
| CN1236302C (en) | Automatic multispectral cell smear and analyzing instrument and method used for cervical cell analysis | |
| US4175860A (en) | Dual resolution method and apparatus for use in automated classification of pap smear and other samples | |
| CA1312276C (en) | Methods and apparatus for immunoploidy analysis | |
| JP6031513B2 (en) | Identification and measurement of reticulocytes | |
| US20170261930A1 (en) | Method and system for detecting and/or classifying cancerous cells in a cell sample | |
| CA1161271A (en) | Method and apparatus for measuring mean cell volume of red blood cells | |
| US5541064A (en) | Methods and apparatus for immunoploidy analysis | |
| CN101339185A (en) | Automatic microscopic imager for detecting cast-off cells and detection method | |
| KR102140385B1 (en) | Cell-zone labeling apparatus and cell-zone detecting system including the same apparatus | |
| CN110363125A (en) | Using the method for Model Transfer identification different cultivars Citrus Huanglongbing pathogen | |
| JP2020169995A (en) | Biosensor platform and method for simultaneous, multiplexed, ultra-sensitive and high throughput optical detection of biomarkers | |
| US20180328848A1 (en) | Cell detection, capture, analysis, aggregation, and output methods and apparatus | |
| US20220383629A1 (en) | Label-free cell classification and screening system based on hybrid transfer learning | |
| CN110335249A (en) | Citrus Huanglongbing pathogen detection method based on high light spectrum image-forming technology | |
| Rodenacker et al. | Automatic analysis of aqueous specimens for phytoplankton structure recognition and population estimation | |
| Chen et al. | Automatic whole blood cell analysis from blood smear using label-free multi-modal imaging with deep neural networks | |
| CN114387596A (en) | Automatic interpretation system for cytopathology smear | |
| CN109214433A (en) | A kind of method that convolutional neural networks distinguish liver cancer differentiation grade | |
| Priyankara et al. | An extensible computer vision application for blood cell recognition and analysis | |
| CA2279945A1 (en) | Infrared spectroscopy for medical imaging | |
| Grimes | Image processing and analysis methods in quantitative endothelial cell biology | |
| Boev et al. | Automated microscopy system for peripheral blood cells |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |