CN109781732A - A kind of small analyte detection and the method for differential counting - Google Patents
A kind of small analyte detection and the method for differential counting Download PDFInfo
- Publication number
- CN109781732A CN109781732A CN201910174311.2A CN201910174311A CN109781732A CN 109781732 A CN109781732 A CN 109781732A CN 201910174311 A CN201910174311 A CN 201910174311A CN 109781732 A CN109781732 A CN 109781732A
- Authority
- CN
- China
- Prior art keywords
- minim
- detection
- original
- differential counting
- machine learning
- 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
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a kind of small analyte detection and the methods of differential counting, this method aims to solve the problem that under the prior art to small analyte detection in the presence of to complicated mixing minim, comprehensively detection and analysis property is poor, it is weak to the character detection property of minim, the technical issues of programmability of determination method is poor, and the digitized samples that can be inquired can not be made；Steps of the method are: detectable substance, which is treated, first with numeralization collecting device carries out orderly digital collection, original digitized data is formed, algorithm of target detection of the detection device based on machine learning and the classification and identification algorithm based on machine learning is recycled to analyze original digitized data.The technical solution passes through the algorithm based on machine learning, it is parallel using acquisition mode and identification method, realize that rapidly respectively different character detect simultaneously differential counting to complicated variety classes minim comprehensively, and the programmability of determination method is improved, and the digitized samples that can be inquired can be made.
Description
Technical field
The invention belongs to the technical field of small analyte detection more particularly to a kind of small analyte detection and the sides of differential counting
Method, it is specific to integrate sensing technology, data sampling and processing analysis, detect to minim the side of simultaneously differential counting
Method.
Background technique
Now, with to the minims such as the microorganism (objects such as cell, microorganism, parasitic ovum, dust particle, pollen particles
Matter) detection technique development, large effect is produced to our life；Currently, being directed to the detection of minim and dividing
Class counts, generally using technologies and laser counting technology, impedance measuring, radio frequencies such as sheath Flow Technique, colorimetric method, flow type cell principles
The technologies such as technology combine, according to impedance value caused by each minim region after testing, forward scattering light, lateral scattering
The signals such as light, lateral fluorescence, to realize the classification to minim and counting.But there is certain to lack for the above method
It falls into, firstly, limitation of the above-mentioned method because of itself, is only capable of working to specific a few class minims, applicability is limited；Its
It is secondary, because respectively there are great differences for character for different types of minim, so if analyzed by above-mentioned method,
This needs to eliminate more distracter, therefore design difficulty is big；Again, above-mentioned method usually requires to form a line minim and suffer
It is a by test point, such detection speed is very slow, inefficiency；Again secondary, above-mentioned method is almost to rely on hardware
Realize detection, and the requirement to hardware is high, and specificity is strong, so that the Setup Cost of detection device is high, simultaneously
Programmability is low, is unfavorable for the utilization of new technology；Finally, above-mentioned method can not provide intuitive digitlization sample for minim
This, inconvenient subsequent digitlization is retained and inquiry.
Summary of the invention
(1) technical problems to be solved
In view of the deficiencies of the prior art, the purpose of the present invention is to provide a kind of small analyte detection and the sides of differential counting
Method, this method, which aims to solve the problem that under the prior art, has to the comprehensive detection and analysis property of complicated mixing minim small analyte detection
Difference, weak to the character detection property of minim, detection speed is slow, and the detection and classification establish to minim are poor, detection and analysis side
The technical issues of programmability of method is poor, and the digitized samples that can be inquired can not be made；The technical solution is by being based on engineering
The algorithm of habit, it is parallel using acquisition mode and identification method, while all targets occurred in watch window are acquired and
Detection realizes that rapidly respectively different character detect simultaneously differential counting to complicated variety classes minim comprehensively, and
The programmability of determination method is improved, and the digitized samples that can be inquired can be made.
(2) technical solution
In order to solve the above-mentioned technical problems, the present invention provides such a small analyte detection and the method for differential counting,
The specific steps of this method are as follows: treat detectable substance using numeralization collecting device and carry out orderly digital collection, formed original
Digitalized data, and obtained original digitized data is transferred to and can be filled to the detection that original digitized data is analyzed
It sets, detection device is recycled to pass through R-CNN, Fast/Faster R-CNN or SSD pairs based on the algorithm of target detection of machine learning
Original digitized data is handled, and obtains all location and range for appearing in the minim in original digitized data, right
In the location and range that each is detected, detection device again using the classification and identification algorithm based on machine learning by SVM or
KNN carries out exact classification to the original digitized data of the location and range, and finds out the corresponding properties and characteristics of all minims,
Minim location information, minim classification information, the minim trait information that finally identification is obtained export as analysis data；
Wherein, the numeralization collecting device be built-in one or more sensors and it is built-in it is one or more be biography
The observation device of sensor offer observation condition.
Preferably, the sensor is CCD sensitive chip or CMOS sensitive chip.
Preferably, the observation device is microscope, narrow planar stripe laser or specific band optical generator.
Preferably, the organizational form of the original digitized data be audio/video flow, network binary stream, audio/video file,
Image file or database file.
Wherein, the observation device of numeralization collecting device and sensor treat detectable substance each time and carry out having for data acquisition
Effect range is referred to as watch window, and targeted transformations all in watch window can be disposably original by the numeralization collecting device
Beginning digitalized data, for acquisition range exceed watch window the case where, numeralization collecting device then can be used repeatedly part adopts
Collection, to complete whole acquisition tasks；Object to be detected can be the mixture of a variety of minims, be also possible to one or more
The mixture of minim and liquid, also, described the functions such as fluorescence or dyeing can be had for mixed liquid；Object to be detected
It can be distributed in a plane, numeralization collecting device can repeatedly be adopted by mobile watch window and the relative position of the plane
Collection, object to be detected also can be distributed in liquid, the numeralization collecting device by the mixed liquor in the form of sufficiently thin plane,
Watch window is flowed through in an orderly manner to be acquired.
Wherein, using identifying the corresponding properties and characteristics of minim, certain judgement effect that you can get it, such as leucocyte
Properties and characteristics Pei Erge-Xiu Te the deformity of deformity, the as hyposegmentation of nucleus of the neutrophil leucocyte in leucocyte, can utilize
Identify that the corresponding properties and characteristics of minim obtain the judgement.
(3) beneficial effect
Compared with prior art, the beneficial effects of the present invention are: technical solution of the present invention solves under the prior art
It is weak to the character detection property of minim to small analyte detection in the presence of to complicated mixing minim, comprehensively detection and analysis property is poor,
Detection speed is slow, and the detection and classification establish to minim are poor, and the programmability of determination method is poor, and can not be made can look into
The technical issues of digitized samples of inquiry；Technical solution of the present invention first treats detectable substance and carries out digital collection, is formed original
Digitalized data, then by the algorithm of target detection based on machine learning and based on the classification and identification algorithm of machine learning, utilize
Acquisition mode and identification method parallel form, while all targets occurred in watch window are acquired and are detected, and
The corresponding properties and characteristics of all minims are found out, it is the minim location information that finally obtains identification, minim classification information, micro-
Small object trait information output is analysis data, to realize comprehensively rapidly respectively different to complicated variety classes minim
Character detect and differential counting, and greatly improve the programmability of determination method, and can be made and can inquire
Digitized samples.
Specific embodiment
To be easy to understand the technical means, the creative features, the aims and the efficiencies achieved by the present invention, below to this
Technical solution in invention specific embodiment carries out clear, complete description, with the present invention is further explained, it is clear that retouched
The specific embodiment stated is only a part of embodiment of the invention, rather than whole patterns.
Embodiment 1
Present embodiment is to carry out identification classification to the smear made and count, and identifies wherein different cells
And parasitic ovum, the specific steps are that: detectable substance is treated using numeralization collecting device and carries out orderly digital collection, is formed
Original digitized data, and obtained original digitized data is transferred to can be to the detection that original digitized data is analyzed
Device recycles detection device to pass through Fast/Faster R-CNN to original figure based on the algorithm of target detection of machine learning
Change data to be handled, all location and range for appearing in the minim in original digitized data is obtained, for each
The location and range detected, detection device again using the classification and identification algorithm based on machine learning by SVM to the position and
The original digitized data of range carries out exact classification, and finds out the corresponding properties and characteristics of all minims, will finally identify
Minim location information out, minim classification information, the output of minim trait information are analysis data.
Wherein, the numeralization collecting device is that in-built CCD sensitive chip and the interior respective sensor that is set to provide observation
The narrow planar stripe laser of condition.
Embodiment 2
Present embodiment is to acquiring and the liquid mixture added with dyeing function carries out identification classification and counts,
Identify wherein allogenic cell various trait, the specific steps are that: detectable substance, which is treated, using numeralization collecting device carries out orderly
Digital collection forms original digitized data, and obtained original digitized data is transferred to can be to raw digitized number
According to the detection device analyzed, detection device is recycled to pass through SSD to original number based on the algorithm of target detection of machine learning
Word data are handled, and all location and range for appearing in the minim in original digitized data are obtained, for each
A location and range detected, detection device pass through KNN to the position using the classification and identification algorithm based on machine learning again
Exact classification is carried out with the original digitized data of range, and finds out the corresponding properties and characteristics of all minims, it finally will identification
Minim location information, minim classification information, the minim trait information obtained exports as analysis data.
Wherein, the numeralization collecting device is that built-in CMOS sensitive chip and the interior respective sensor that is set to provide sight
Examine the microscope of condition.
The foregoing describe technical characteristics of the invention and basic principle and associated advantages, for those skilled in the art
For, it is clear that the present invention is not limited to the details of above-mentioned exemplary embodiment, and without departing substantially from design of the invention or
In the case where essential characteristic, the present invention can be realized in other specific forms.Therefore, in all respects, should all incite somebody to action
Above-mentioned specific embodiment regards exemplary as, and is non-limiting, the scope of the present invention by appended claims and
It is not that above description limits, it is intended that all changes that come within the meaning and range of equivalency of the claims are included
In the present invention.
Although not each embodiment is only in addition, it should be understood that this specification is described according to each embodiment
It contains an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art answer
When considering the specification as a whole, the technical solution in each embodiment may also be suitably combined to form art technology
The other embodiments that personnel are understood that.
Claims (4)
1. a kind of small analyte detection and the method for differential counting, which is characterized in that the specific steps of this method are as follows: utilize digitlization
Acquisition device treats detectable substance and carries out orderly digital collection, forms original digitized data, and the original figure that will be obtained
Change data and is transferred to and detection device can be recycled based on machine learning the detection device that original digitized data is analyzed
Algorithm of target detection is handled original digitized data by R-CNN, Fast/Faster R-CNN or SSD, is obtained all
The location and range for appearing in the minim in original digitized data, for the location and range that each is detected, detection
Device passes through SVM or KNN to the raw digitized number of the location and range using the classification and identification algorithm based on machine learning again
According to exact classification is carried out, and the corresponding properties and characteristics of all minims are found out, the minim location information that finally obtains identification,
Minim classification information, the output of minim trait information are analysis data；
Wherein, the numeralization collecting device be built-in one or more sensors and it is built-in it is one or more be sensor
The observation device of observation condition is provided.
2. a kind of small analyte detection according to claim 1 and the method for differential counting, which is characterized in that the sensor
For CCD sensitive chip or CMOS sensitive chip.
3. a kind of small analyte detection according to claim 1 and the method for differential counting, which is characterized in that the observation dress
It is set to microscope, narrow planar stripe laser or specific band optical generator.
4. a kind of small analyte detection according to claim 1 and the method for differential counting, which is characterized in that the original number
The organizational form of word data is audio/video flow, network binary stream, audio/video file, image file or database file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910174311.2A CN109781732A (en) | 2019-03-08 | 2019-03-08 | A kind of small analyte detection and the method for differential counting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910174311.2A CN109781732A (en) | 2019-03-08 | 2019-03-08 | A kind of small analyte detection and the method for differential counting |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109781732A true CN109781732A (en) | 2019-05-21 |
Family
ID=66487689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910174311.2A Pending CN109781732A (en) | 2019-03-08 | 2019-03-08 | A kind of small analyte detection and the method for differential counting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109781732A (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021990A (en) * | 2016-06-07 | 2016-10-12 | 广州麦仑信息科技有限公司 | Method for achieving classification and self-recognition of biological genes by means of specific characters |
| CN106096561A (en) * | 2016-06-16 | 2016-11-09 | 重庆邮电大学 | Infrared pedestrian detection method based on image block degree of depth learning characteristic |
| CN106780498A (en) * | 2016-11-30 | 2017-05-31 | 南京信息工程大学 | Based on point depth convolutional network epithelium and matrix organization's automatic division method pixel-by-pixel |
| CN106845401A (en) * | 2017-01-20 | 2017-06-13 | 中国科学院合肥物质科学研究院 | A kind of insect image-recognizing method based on many spatial convoluted neutral nets |
| CN106886795A (en) * | 2017-02-17 | 2017-06-23 | 北京维弦科技有限责任公司 | Object identification method based on the obvious object in image |
| US20170236183A1 (en) * | 2016-02-11 | 2017-08-17 | Ebay Inc. | System and method for detecting visually similar items |
| CN107169556A (en) * | 2017-05-15 | 2017-09-15 | 电子科技大学 | stem cell automatic counting method based on deep learning |
| CN107301640A (en) * | 2017-06-19 | 2017-10-27 | 太原理工大学 | A kind of method that target detection based on convolutional neural networks realizes small pulmonary nodules detection |
| CN107341506A (en) * | 2017-06-12 | 2017-11-10 | 华南理工大学 | A kind of Image emotional semantic classification method based on the expression of many-sided deep learning |
| CN108021936A (en) * | 2017-11-28 | 2018-05-11 | 天津大学 | A kind of tumor of breast sorting algorithm based on convolutional neural networks VGG16 |
| CN108335298A (en) * | 2018-03-20 | 2018-07-27 | 东南大学 | Cereal-granules counting device |
| CN108985322A (en) * | 2018-06-01 | 2018-12-11 | 广东电网有限责任公司 | A kind of cable tunnel ponding positioning identifying method based on ZF-Faster RCNNs |
| CN109117826A (en) * | 2018-09-05 | 2019-01-01 | 湖南科技大学 | A kind of vehicle identification method of multiple features fusion |
| CN109190622A (en) * | 2018-09-11 | 2019-01-11 | 深圳辉煌耀强科技有限公司 | Epithelial cell categorizing system and method based on strong feature and neural network |
| CN109344859A (en) * | 2018-08-19 | 2019-02-15 | 天津大学 | A kind of mitotic mapping based on incorporation time pond operator and knowledge method for distinguishing |
-
2019
- 2019-03-08 CN CN201910174311.2A patent/CN109781732A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170236183A1 (en) * | 2016-02-11 | 2017-08-17 | Ebay Inc. | System and method for detecting visually similar items |
| CN106021990A (en) * | 2016-06-07 | 2016-10-12 | 广州麦仑信息科技有限公司 | Method for achieving classification and self-recognition of biological genes by means of specific characters |
| CN106096561A (en) * | 2016-06-16 | 2016-11-09 | 重庆邮电大学 | Infrared pedestrian detection method based on image block degree of depth learning characteristic |
| CN106780498A (en) * | 2016-11-30 | 2017-05-31 | 南京信息工程大学 | Based on point depth convolutional network epithelium and matrix organization's automatic division method pixel-by-pixel |
| CN106845401A (en) * | 2017-01-20 | 2017-06-13 | 中国科学院合肥物质科学研究院 | A kind of insect image-recognizing method based on many spatial convoluted neutral nets |
| CN106886795A (en) * | 2017-02-17 | 2017-06-23 | 北京维弦科技有限责任公司 | Object identification method based on the obvious object in image |
| CN107169556A (en) * | 2017-05-15 | 2017-09-15 | 电子科技大学 | stem cell automatic counting method based on deep learning |
| CN107341506A (en) * | 2017-06-12 | 2017-11-10 | 华南理工大学 | A kind of Image emotional semantic classification method based on the expression of many-sided deep learning |
| CN107301640A (en) * | 2017-06-19 | 2017-10-27 | 太原理工大学 | A kind of method that target detection based on convolutional neural networks realizes small pulmonary nodules detection |
| CN108021936A (en) * | 2017-11-28 | 2018-05-11 | 天津大学 | A kind of tumor of breast sorting algorithm based on convolutional neural networks VGG16 |
| CN108335298A (en) * | 2018-03-20 | 2018-07-27 | 东南大学 | Cereal-granules counting device |
| CN108985322A (en) * | 2018-06-01 | 2018-12-11 | 广东电网有限责任公司 | A kind of cable tunnel ponding positioning identifying method based on ZF-Faster RCNNs |
| CN109344859A (en) * | 2018-08-19 | 2019-02-15 | 天津大学 | A kind of mitotic mapping based on incorporation time pond operator and knowledge method for distinguishing |
| CN109117826A (en) * | 2018-09-05 | 2019-01-01 | 湖南科技大学 | A kind of vehicle identification method of multiple features fusion |
| CN109190622A (en) * | 2018-09-11 | 2019-01-11 | 深圳辉煌耀强科技有限公司 | Epithelial cell categorizing system and method based on strong feature and neural network |
Non-Patent Citations (1)
| Title |
|---|
| 刘树杰: "基于卷积神经网络的红细胞检测和计数方法", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Holt et al. | Principles and methods for automated palynology | |
| Jayakody et al. | Microscope image based fully automated stomata detection and pore measurement method for grapevines | |
| CN103890561B (en) | The optical detection of particle and analysis | |
| CN105940301B (en) | A kind of stream type cell analyzer and its multidimensional data sorting technique, device | |
| US9772282B2 (en) | System for wide field-of-view, highly oblique illumination microscopy for scatter-based discrimination of cells | |
| US20170052106A1 (en) | Method for label-free image cytometry | |
| US10337975B2 (en) | Method and system for characterizing particles using a flow cytometer | |
| CN103984939B (en) | A kind of sample visible component sorting technique and system | |
| CN105181649B (en) | A kind of Novel free marking mode identifies cell instrument method | |
| Kezlarian et al. | Artificial intelligence in thyroid fine needle aspiration biopsies | |
| CN103345654A (en) | Method for differential counting of white blood cells based on morphology | |
| CN105043998A (en) | Method for identifying corn haploid | |
| Lyashenko et al. | Wavelet Analysis of Cytological Preparations Image in Different Color Systems | |
| Kajtár et al. | Automated fluorescent in situ hybridization (FISH) analysis of t (9; 22)(q34; q11) in interphase nuclei | |
| Deshmukh et al. | A confirmatory test for sperm in sexual assault samples using a microfluidic-integrated cell phone imaging system | |
| Sandmann et al. | Multidimensional single-cell analysis based on fluorescence microscopy and automated image analysis | |
| CN104751188B (en) | Picture processing method and system | |
| Di Ruberto et al. | A multiple classifier learning by sampling system for white blood cells segmentation | |
| Fuda et al. | Artificial intelligence in clinical multiparameter flow cytometry and mass cytometry–key tools and progress | |
| CN109580550A (en) | A kind of classification processing method and its device of leucocyte | |
| CN101231229A (en) | Non-dyeing automatic counting method for liquid bacterium-containing quantity | |
| CN109781732A (en) | A kind of small analyte detection and the method for differential counting | |
| Di Ruberto et al. | Learning by sampling for white blood cells segmentation | |
| CN102187334A (en) | Shape parameter for hematology instruments | |
| CN113380318B (en) | Artificial intelligence assisted flow cytometry 40CD immunophenotyping detection method and system |
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: 20190521 |
|
| RJ01 | Rejection of invention patent application after publication |