CN109242842A - Human chromosomal analytical equipment, equipment and storage medium based on image recognition - Google Patents
Human chromosomal analytical equipment, equipment and storage medium based on image recognition Download PDFInfo
- Publication number
- CN109242842A CN109242842A CN201811013795.4A CN201811013795A CN109242842A CN 109242842 A CN109242842 A CN 109242842A CN 201811013795 A CN201811013795 A CN 201811013795A CN 109242842 A CN109242842 A CN 109242842A
- Authority
- CN
- China
- Prior art keywords
- chromosome
- image
- independent
- independent chromosome
- analytical equipment
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses the human chromosomal analytical equipments based on image recognition, image corresponding with chromosome is obtained by acquisition module first, then image procossing is carried out to image by processing module and obtains target image, and the chromosome in target image is split, obtain a plurality of independent chromosome, extraction module carries out feature extraction to each independent chromosome, karyological character corresponding with each independent chromosome is obtained, is finally classified each independent chromosome to analyze each independent chromosome using trained classifier by analysis module.Therefore, the application can analyze chromosome using intelligentized mode, compared with the prior art in by the way of artificial for, the a large amount of time can not only be saved by the way of automation, nor limited by technical experience, greatly enhance the analysis efficiency and reliability to chromosome.In addition, the invention also discloses a kind of analytical equipment and storage medium for chromosome, effect is as above.
Description
Technical field
The present invention relates to medical domain, in particular to a kind of human chromosomal analytical equipment based on image recognition, equipment
And computer storage medium.
Background technique
Chromosome analysis is an important conventional detection means in cytogenetics, and is the diagnosis for detecting chromosomal disorders
Foundation.Analysis detection is carried out mainly to fragment deletion, repetition, the inversion in chromosome, displacement and chromosome number to chromosome
The detection of mesh variation etc..
The conventional method analyzed at present chromosome is to be analyzed using artificial method chromosome, with determination
Chromosome and its relevant disease, can Mao to the analysis of every dyad one by one due to including 23 pairs of chromosomes in human body
Take technical staff's a large amount of time, and is affected by subjective factor.Such as technical experience due to technical staff and energy have
Limit, encounters the chromosomal abnormalities that can not be distinguished, needs to take longer time and analyze and research, or because that can not distinguish dyeing
Body abnormal conditions and cause to judge by accident, reliability is lower.
Therefore, the technical bottleneck that the efficiency and reliability analyzed chromosome is this field how is improved.
Summary of the invention
It is an object of the invention to disclose a kind of human chromosomal analytical equipment based on image recognition, equipment and storage to be situated between
Matter improves the efficiency and reliability that chromosome is analyzed.
To achieve the above object, the embodiment of the invention discloses following technical solutions:
First, the embodiment of the invention discloses a kind of human chromosomal analytical equipment based on image recognition, comprising:
Module is obtained, for obtaining image corresponding with chromosome;
Processing module obtains target image for carrying out image procossing to described image, and in the target image
Chromosome is split, and obtains a plurality of independent chromosome;
Extraction module obtains and each independent chromosome pair for carrying out feature extraction to each independent chromosome
The karyological character answered;
Analysis module, for utilizing trained classification according to karyological character corresponding with each independent chromosome
Device classifies to analyze each independent chromosome to each independent chromosome.
Preferably, the processing module specifically includes:
First processing units, for carrying out image filtering processing to described image to remove the noise in described image;
The second processing unit, for through the first processing units treated image carry out image binaryzation processing with
Described image is converted into artwork master;
Third processing unit, for being split based on Contour following algorithm pair artwork master corresponding with the chromosome
Obtain the target image comprising chromosome;
Fourth processing unit, for being split based on morphological images processing to the chromosome in the target image,
Obtain a plurality of independent chromosome.
Preferably, the fourth processing unit specifically includes:
Handle subelement, for identification in the target image chromosome boundary line, according to the boundary of the chromosome
The chromosome is divided into a plurality of independent chromosome by line.
Preferably, the extraction module specifically includes:
First extraction unit is extracted for the central axes to each independent chromosome;
Second extraction unit is extracted for the chromosome length to each independent chromosome;
Third extraction unit is extracted for the centromere index to each independent chromosome;
4th extraction unit extracts, wherein the karyological character for the banding pattern to each independent chromosome
Including the central axes, the chromosome length, the centromere index and the banding pattern.
Preferably, the extraction module further include:
5th extraction unit is extracted for the area to each independent chromosome.
Preferably, the analysis module includes:
Classifier training unit, for by input sample data corresponding with the chromosome to preliminary classification device into
Row training, obtains training classifier, the sample data include normal chromosomal and normal chromosomal polymorphisms characteristic data and
Abnormal chromosome characteristic;
Taxon, for being carried out using the trained classifier and the karyological character to each independent chromosome
Classification, exports multiple classification results;
Analytical unit is analyzed for the output result to the trained classifier.
Second, the human chromosomal analytical equipment based on image recognition that the embodiment of the invention discloses a kind of, comprising:
Memory, for storing computer program;
Processor when for executing the computer program stored in the memory, realizes following steps:
Obtain image corresponding with chromosome;
Image procossing is carried out to described image and obtains target image, and the chromosome in the target image is divided
It cuts, obtains a plurality of independent chromosome;
Feature extraction is carried out to each independent chromosome, it is special to obtain chromosome corresponding with each independent chromosome
Sign;
According to karyological character corresponding with each independent chromosome, using trained classifier to each independence
Chromosome is classified to analyze each independent chromosome.
Third is stored with computer program on the computer readable storage medium, and the computer program is by processor
When execution, following steps are realized:
Obtain image corresponding with chromosome;
Image procossing is carried out to described image and obtains target image, and the chromosome in the target image is divided
It cuts, obtains a plurality of independent chromosome;
Feature extraction is carried out to each independent chromosome, it is special to obtain chromosome corresponding with each independent chromosome
Sign;
According to karyological character corresponding with each independent chromosome, using trained classifier to each independence
Chromosome is classified to analyze each independent chromosome.
As it can be seen that a kind of human chromosomal analytical equipment based on image recognition disclosed by the embodiments of the present invention, first by obtaining
Modulus block obtains image corresponding with chromosome, then carries out image procossing to image by processing module and obtains target image, and
Chromosome in target image is split, a plurality of independent chromosome is obtained, extraction module carries out each independent chromosome special
Sign is extracted, and karyological character corresponding with each independent chromosome is obtained, finally by analysis module according to each independent chromosome pair
The karyological character answered classifies to divide each independent chromosome to each independent chromosome using trained classifier
Analysis.Therefore, by analytical equipment disclosed in the present application, human chromosomal can be analyzed by the way of automation, phase
Than a large amount of time can not only be saved by the way of automation in the prior art by the way of artificial, and
It is not limited by technical experience, greatly enhances the analysis efficiency and reliability to chromosome.In addition, the present invention is implemented
Example also discloses a kind of analytical equipment and storage medium for chromosome, and effect is as above.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of human chromosomal analytical equipment structural representation based on image recognition disclosed by the embodiments of the present invention
Figure;
Fig. 2 is a kind of human chromosomal analytical equipment structural representation based on image recognition disclosed by the embodiments of the present invention
Figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of, and human chromosomal analytical equipment, equipment and storage based on image recognition are situated between
Matter improves the efficiency and reliability that chromosome is analyzed.
Referring to Figure 1, Fig. 1 is a kind of human chromosomal analytical equipment based on image recognition disclosed by the embodiments of the present invention
Structural schematic diagram, the analytical equipment include:
Module 101 is obtained, for obtaining image corresponding with chromosome.
Processing module 102 obtains target image for carrying out image procossing to image, and to the chromosome in target image
It is split, obtains a plurality of independent chromosome.
Extraction module 103 obtains dye corresponding with each independent chromosome for carrying out feature extraction to each independent chromosome
Colour solid feature.
Analysis module 104, for utilizing trained classifier according to karyological character corresponding with each independent chromosome
Classify each independent chromosome to analyze each independent chromosome.
Specifically, the image obtained in module 101 is by Zeiss microscope or to come the micro- scarnning mirror of card in the present embodiment
(image is to be observed the chromosome slide of human body cell, screened to obtain ideal dye under the microscope to lower machine image
Colour solid image, ideal chromosome image refer to it is microscopical within sweep of the eye, genome (46 independent chromosome)
It is high-visible), all high-visible chromosome of 23 pairs of chromosomes within the scope of field of microscope is carried out by photography using camera and is taken
According to, and using the picture of shooting as the image in the embodiment of the present invention.It is by which kind of microscope or electronic device about image
It obtains, the embodiment of the present invention is simultaneously not construed as limiting.
Further, the main purpose of processing module 102 is to be split to obtain 46 by 23 pairs of chromosomes (46 chromosome)
The independent chromosome of item, the process of processing is: first pre-processing image, i.e., by the image-region comprising chromosome from image
In extract, pretreated key link include: image filtering first is carried out to image, greyscale transformation, wavelet transformation etc. will figure
As carrying out clarity processing to obtain the area image and the biggish image of background image discrimination comprising chromosome.Followed by scheme
As partitioning portion, image segmentation part mainly includes two big processes, first stage be to the image-region comprising chromosome with
Image background separates.It, will be in the target image comprising chromosomal region after obtaining the target image comprising chromosomal region
Chromosome is split to obtain 46 independent chromosomes, and the main process of image segmentation is: first to the original comprising 46 chromosome
Beginning image carries out gradient binaryzation, and then is based on wheel to the original image for carrying out gradient binaryzation based on Contour following algorithm
Profile track algorithm separates chromosomal region and background image to obtain the target image comprising chromosome, is finally based on form
It learns image procossing to be split the chromosome in target image, obtains a plurality of independent chromosome.About the detailed of processing module 102
Thin process will describe in detail in next embodiment.
Secondly, in the present embodiment, by taking 46 chromosome as an example, obtaining 46 independent chromosomes about extraction module 103
Later, every independent chromosome all includes tinction body characteristics, and the type of karyological character mainly includes following several: chromosome
Length, centromere index, area, central axes and banding pattern etc..It should be noted that being mentioned carrying out feature to every independent chromosome
When taking, can only extract can most react the feature of chromosome information, it is not necessary to by all feature extractions relevant to chromosome,
Such as using chromosome length, centromere index and banding pattern as the karyological character of each independent chromosome.
For analysis module 104, after the karyological character for obtaining each independent chromosome, as the defeated of classifier
Enter, using the classification results of the corresponding chromosome of each karyological character as output, for the part, it is necessary first to initial point
Class device is trained, and the sample data that inputs includes: that (each chromosome has respectively right for the serial number of 46 chromosome when training
The serial number answered), (including chromosome length, abnormal, chromosome banding pattern is abnormal for the exception information of the corresponding chromosome of each serial number
Deng), normal information (the normal banding pattern of normal length information and chromosome including chromosome of the corresponding chromosome of each serial number
Information etc.), after sample data is input to classifier, the classification results (normal dyeing of 46 chromosome is exported by classifier
The classification and arrangement of body, the classification of abnormal chromosome and arrangement and type and name (chromosome repetition, insertion, transposition, inversion
And numerical abnormality) etc. classification results).
As it can be seen that a kind of human chromosomal analytical equipment based on image recognition disclosed by the embodiments of the present invention, first by obtaining
Modulus block obtains image corresponding with chromosome, then carries out image procossing to image by processing module and obtains target image, and
Chromosome in target image is split, a plurality of independent chromosome is obtained, extraction module carries out each independent chromosome special
Sign is extracted, and karyological character corresponding with each independent chromosome is obtained, finally by analysis module according to each independent chromosome pair
The karyological character answered is classified and is analyzed to each independent chromosome using trained classifier.Therefore, pass through the application
Disclosed analytical equipment can analyze chromosome using intelligentized mode, middle using artificial compared with the prior art
Mode for, a large amount of time can not only be saved by the way of automation, nor limited by technical experience, from one
Determine to improve the analysis efficiency and reliability to chromosome in degree.
Based on the above embodiment, as preferred embodiment, processing module 102 includes:
First processing units, for carrying out image filtering processing to image to remove the noise in image.
The second processing unit, for carrying out image binaryzation processing to the image after first processing units to turn image
It is changed to artwork master.
Third processing unit, for being split to obtain based on Contour following algorithm artwork master corresponding for chromosome
Target image comprising chromosome.
Fourth processing unit is obtained for being split based on morphological images processing to the chromosome in target image
A plurality of independent chromosome.
Specifically, may refer to by first processing units to image progress image filtering processing existing in the present embodiment
Technology, mainly carrying out noise reduction or denoising, denoising and noise reduction for image to the image comprising chromosome can adopt
With following three kinds of methods: the first, it can be in frequency domain using the high band in low pass filtering method removal noise frequency range;Second,
Noise reduction and denoising can be carried out using the field method of average and median filtering method in airspace.It is flat about low pass filtering method denoising, field
Equal method denoising and median filtering method denoising may refer to the prior art.Details are not described herein for the embodiment of the present invention.
The second processing unit be used for by denoising after image carry out binary conversion treatment, the purpose of binary conversion treatment be by
Image after denoising is converted to artwork master (can also carry out gray processing to the image after denoising), carries out two to the image after denoising
The purpose of value processing is: increasing the contrast between chromosome and background.The difference for reducing oolemma and blanking bar inside chromosome, from
And be conducive to as a whole separate chromosome from background.
Third processing unit is used to be split the corresponding artwork master of chromosome based on Contour following algorithm and be wrapped
Target image containing chromosome.Wherein, Contour following algorithm may refer to the prior art, be applied in the embodiment of the present invention,
After obtaining the biggish artwork master of contrast between chromosome and background, the boundary between chromosome and background is apparent, chromosome
The boundary of internal bright band and blanking bar is also clear that.Based on this, whole chromosome region is determined most using Contour following algorithm
The boundary of periphery, also just obtains the boundary between chromosomal region and background.The purpose of Contour following algorithm is along figure
The boundary search in equal colors region, the point on the boundary line searched (profile) is recorded in point range.Each point range is meant that
One contour line.In the embodiment of the present invention, the pixel of finally obtained chromosome is white, and the pixel of background colour is black.Then
The step of boundary search, is as follows: being to search for according to graph scanning direction first, checks that the pixel of present image is white or black, general
Starting point of the black pixel detected at first as Contour tracking, is started with the starting point, detect in sequence each pixel whether be
Black pixel stops tracking if each pixel detected in sequence is white, and the contour line of this secondary tracking is made
To track first contour line completed, and so on.Certainly, based on contour following algorithm to artwork master carry out Contour extraction from
And the region part comprising chromosome and background parts are subjected to the isolated area image comprising chromosome.
In fourth processing unit, the chromosome in target image is split based on morphological images processing, is obtained more
Item independence chromosome.About this part, the concept of morphological images processing may refer to the prior art, is applied to the present invention and implements
In example, after obtaining the area image comprising chromosome, scanned then along the X-axis of the area image, when discovery and background colour
When different pixel, is scanned for along the boundary of the pixel, obtain the boundary of this chromosome, obtain item chromosome
Afterwards, its boundary information is recorded, the obtained chromosome is erased from background, then proceedes to extract remaining chromosome, according to
It is secondary to extract 46 independent chromosomes.
Wherein, as preferred embodiment, fourth processing unit specifically includes processing subelement, for identification target image
Chromosome is divided into a plurality of independent chromosome according to the boundary line of chromosome by the boundary line of middle chromosome.
Based on the above embodiment, extraction module 103 specifically includes:
First extraction unit is extracted for the central axes to each independent chromosome.
Second extraction unit is extracted for the chromosome length to each independent chromosome.
Third extraction unit is extracted for the centromere index to each independent chromosome.
4th extraction unit extracts, wherein karyological character includes axis for the banding pattern to each independent chromosome
Line, chromosome length, centromere index and banding pattern.
Specifically, the first extraction unit extracts chromosome axis from each independent chromosome, then root in the present embodiment
The length of each independent chromosome, centromere index and banding pattern are determined according to the independent chromosome axis index extracted.Wherein, smart
The true axis for extracting chromosome is most important to segmentation Cross reaction body, calculating chromosome length, searching centromere position.
Wherein, the axis in independent chromosome is extracted including following two method: inscribe circule method and refinement method, about inscribed circle
Method is also referred to as skeleton method, and key step is as follows: the nearest neighbor point on zone boundary is found out to the every bit in independent chromosome, if
Certain point has more than one boundary point to have identical minimum range with it in the region of the independent chromosome of this, then the point is located at symmetrical
Then the adjacent axial point found out is connected into skeleton by axis or axis.In this way, the skeleton is just the independent chromosome found out
Axis.Method is followed by refined, refinement method is continuously to remove company of the outermost layer element until acquisition unit width of independent chromosome
Logical line (as central axes), therefore thinning algorithm is exactly the boundary pixel for repeating to peel off two-value independence chromosome.It is calculated about refinement
Method may refer to the prior art, be applied in the embodiment of the present invention, refinement step are as follows: determine the outermost side of independent chromosome
Then the boundary point is assigned a value of the background value of independent chromosome by boundary, that is, independent chromosome is reduced a circle.Successively
Analogize, the skeleton (i.e. central axes) of each independent chromosome can be obtained.
Second extraction unit is the extraction to chromosome length, which can be in for the extraction of chromosome length
The length of axis is indicated, and therefore, the acquisition of chromosome length is exactly the central axes measurement of length to independent chromosome.Solely
Vertical chromosome map seems to be stored in a manner of bitmap, that is to say, that the central axes of independent chromosome are also by some connected point sets
Composition, the axis line process for measuring independent chromosome by these connected point sets are as follows: finding chromosome central axes first
One endpoint, and the pixel of the endpoint is recorded, remember that length at this time is 0, advances since the endpoint along central axes, take second
It is a, the relationship of the endpoint and second point is recorded, if second point is located in the first neighborhood of a point, remembers that its relationship is straight line phase
Neighbour, length increases by 1.1, if second point is located on the tilted direction of endpoint, remembers that its relationship is that oblique line is adjacent, length increase by 1.5 (according to
Actual conditions, the numerical value at this can change).It repeats the above process, another end until promoting independent chromosome central axes
Until point, the length of measured central axes is the length of independent chromosome central axes at this time.
Third unit has been obtained about the extraction of independent chromosome centromere index by the processing of above each step
Then the contour curve of each independent chromosome and central axes take vertical line to the point on central axes, at this point, vertical line and contour line phase
Friendship obtains intersection point, obtains the vertical line section by the point on the intersection point and central axes, finds out hanging down for every bit on chromosome central axes
The length of line segment is the position where centromere corresponding to the point on the central axes where most short vertical line section.
4th extraction unit is extracted to the banding pattern of independent chromosome, and the banding pattern expression of chromosome is chromosome map
It, can be directly using the gray value of each point on chromosome central axes as chromosome banding pattern as the intensity profile along central axes.This
In inventive embodiments, to each point on chromosome central axes, the average ash of each target point on its vertical line relative to central axes is taken
Band offset of the angle value as the point.
Wherein, as preferred embodiment, extraction module 103 further include:
5th extraction unit is extracted for the area to each independent chromosome.
Specifically, extracting each boundary for solving independent chromosome to the area of independent chromosome in the present embodiment
The area of the closed area surrounded.
Based on the above embodiment, as preferred embodiment, analysis module 104 includes: classifier training unit, for leading to
The sample data corresponding with chromosome for crossing input is trained preliminary classification device, obtains training classifier, sample data packet
Include normal chromosomal characteristic and abnormal chromosome characteristic.
Taxon is exported more for being classified using training classifier and karyological character to each independent chromosome
A classification results.
Analytical unit is analyzed for the output result to training classifier.
Specifically, in the present embodiment, after the karyological character for obtaining independent chromosome, using karyological character as classification
The input of device, each independence chromosome all have corresponding chromosome serial number.By the dyeing of each independent chromosome
Body characteristics input classifier after, classifier be just automatically learned normal chromosomal under each serial number, abnormal chromosome and
The type of abnormal chromosome.In the embodiment of the present invention, it can be classified using geometric classfication method, it can also be using manual point
Class.For geometric classfication method, first independent chromosome can be classified according to long end and short end, and then independently be dyed each
Body is divided into the genome of different length classification.Secondly, again with centromere be according to classify chromosome is divided into it is different
Independent chromosome is finally divided into different banding pattern groups based on banding pattern again by centromere group.Hereafter, then to the independence in each group classification
Chromosome is reset and is grouped again.To which classification results are determined as numerical abnormalities of chromosomes, chromosome translocation, chromosome are fallen
The classifications such as position and chromosome deficiency insertion.
A kind of human chromosomal analytical equipment based on image recognition disclosed by the embodiments of the present invention is introduced below,
Fig. 2 is referred to, Fig. 2 is a kind of human chromosomal analytical equipment structural representation based on image recognition disclosed by the embodiments of the present invention
Figure, comprising:
Memory 201, for storing computer program;
Processor 202 when for executing the computer program stored in memory, realizes following steps:
Obtain image corresponding with chromosome;
Image procossing is carried out to image and obtains target image, and the chromosome in target image is split, is obtained more
Item independence chromosome;
Feature extraction is carried out to each independent chromosome, obtains karyological character corresponding with each independent chromosome;
According to karyological character corresponding with each independent chromosome, using trained classifier to each independent chromosome into
Row classification is to analyze each independent chromosome.
As it can be seen that a kind of human chromosomal analytical equipment based on image recognition disclosed in the present embodiment, due to that can pass through
Processor calls the analysis program identical with the computer program in the analytical equipment for chromosome of memory storage, thus
The analytical procedure as possessed by human chromosomal analytical equipment of the above-mentioned any embodiment based on image recognition is realized, so well-behaved
Desorption device has with above-mentioned analytical equipment same actual effect.
This programme in order to better understand is stored on a kind of computer readable storage medium disclosed by the embodiments of the present invention
Computer program when computer program is executed by processor, realizes following steps:
Obtain image corresponding with chromosome;
Image procossing is carried out to image and obtains target image, and the chromosome in target image is split, is obtained more
Item independence chromosome;
Feature extraction is carried out to each independent chromosome, obtains karyological character corresponding with each independent chromosome;
According to karyological character corresponding with each independent chromosome, using trained classifier to each independent chromosome into
Row classification is to analyze each independent chromosome.
As it can be seen that a kind of computer readable storage medium disclosed in the present embodiment, is calculated due to that can be called by processor
The analysis program identical with the computer program in the analytical equipment for chromosome of machine readable storage medium storing program for executing storage, thus real
Analytical procedure possessed by the analytical equipment for chromosome now mentioned such as above-mentioned any embodiment, so this analytical equipment has
There is the same actual effect with above-mentioned analytical equipment.
It is situated between above to a kind of human chromosomal analytical equipment based on image recognition of the disclosure as set forth herein, equipment and storage
Matter is described in detail.Specific examples are used herein to illustrate the principle and implementation manner of the present application, above
The explanation of embodiment is merely used to help understand the present processes and its core concept.It should be pointed out that for the art
Those of ordinary skill for, under the premise of not departing from the application principle, can also to the application carry out it is several improvement and repair
Decorations, these improvement and modification are also fallen into the protection scope of the claim of this application.
Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with other realities
The difference of example is applied, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part illustration
?.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (8)
1. a kind of human chromosomal analytical equipment based on image recognition characterized by comprising
Module is obtained, for obtaining image corresponding with chromosome;
Processing module obtains target image for carrying out image procossing to described image, and to the dyeing in the target image
Body is split, and obtains a plurality of independent chromosome;
Extraction module obtains corresponding with each independent chromosome for carrying out feature extraction to each independent chromosome
Karyological character;
Analysis module, for utilizing trained classifier pair according to karyological character corresponding with each independent chromosome
Each independent chromosome is classified to analyze each independent chromosome.
2. the human chromosomal analytical equipment according to claim 1 based on image recognition, which is characterized in that the processing
Module specifically includes:
First processing units, for carrying out image filtering processing to described image to remove the noise in described image;
The second processing unit, for carrying out image binaryzation processing to through the first processing units treated image with by institute
It states image and is converted to artwork master;
Third processing unit, for being split to obtain based on Contour following algorithm pair artwork master corresponding with the chromosome
Target image comprising chromosome;
Fourth processing unit is obtained for being split based on morphological images processing to the chromosome in the target image
The a plurality of independent chromosome.
3. the human chromosomal analytical equipment according to claim 2 based on image recognition, which is characterized in that the described 4th
Processing unit specifically includes:
Handle subelement, for identification in the target image chromosome boundary line, according to the boundary line of the chromosome will
The chromosome is divided into a plurality of independent chromosome.
4. the human chromosomal analytical equipment according to claim 1 based on image recognition, which is characterized in that the extraction
Module specifically includes:
First extraction unit is extracted for the central axes to each independent chromosome;
Second extraction unit is extracted for the chromosome length to each independent chromosome;
Third extraction unit is extracted for the centromere index to each independent chromosome;
4th extraction unit is extracted for the banding pattern to each independent chromosome, wherein the karyological character includes
The central axes, the chromosome length, the centromere index and the banding pattern.
5. the human chromosomal analytical equipment according to claim 4 based on image recognition, which is characterized in that the extraction
Module further include:
5th extraction unit is extracted for the area to each independent chromosome.
6. the human chromosomal analytical equipment according to claim 1 based on image recognition, which is characterized in that the analysis
Module includes:
Classifier training unit instructs preliminary classification device for the sample data corresponding with the chromosome by inputting
Practice, obtains training classifier, the sample data includes normal chromosomal and normal chromosomal polymorphisms characteristic data and exception
Karyological character data;
Taxon, for being divided using the trained classifier and the karyological character each independent chromosome
Class exports multiple classification results;
Analytical unit is analyzed for the output result to the trained classifier.
7. a kind of human chromosomal analytical equipment based on image recognition characterized by comprising
Memory, for storing computer program;
Processor when for executing the computer program stored in the memory, realizes following steps:
Obtain image corresponding with chromosome;
Image procossing is carried out to described image and obtains target image, and the chromosome in the target image is split, is obtained
To a plurality of independent chromosome;
Feature extraction is carried out to each independent chromosome, obtains karyological character corresponding with each independent chromosome;
According to karyological character corresponding with each independent chromosome, described independent dyed using trained classifier to each
Body is classified to analyze each independent chromosome.
8. a kind of computer readable storage medium, computer program, feature are stored on the computer readable storage medium
It is, when the computer program is executed by processor, realizes following steps:
Obtain image corresponding with chromosome;
Image procossing is carried out to described image and obtains target image, and the chromosome in the target image is split, is obtained
To a plurality of independent chromosome;
Feature extraction is carried out to each independent chromosome, obtains karyological character corresponding with each independent chromosome;
According to karyological character corresponding with each independent chromosome, described independent dyed using trained classifier to each
Body is classified to analyze each independent chromosome.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811013795.4A CN109242842A (en) | 2018-08-31 | 2018-08-31 | Human chromosomal analytical equipment, equipment and storage medium based on image recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811013795.4A CN109242842A (en) | 2018-08-31 | 2018-08-31 | Human chromosomal analytical equipment, equipment and storage medium based on image recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109242842A true CN109242842A (en) | 2019-01-18 |
Family
ID=65059959
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811013795.4A Pending CN109242842A (en) | 2018-08-31 | 2018-08-31 | Human chromosomal analytical equipment, equipment and storage medium based on image recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242842A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111209900A (en) * | 2020-04-20 | 2020-05-29 | 湖南自兴智慧医疗科技有限公司 | Image processing method for positioning centromere according to feature extraction algorithm |
CN111986183A (en) * | 2020-08-25 | 2020-11-24 | 中国科学院长春光学精密机械与物理研究所 | Chromosome scattergram image automatic segmentation and identification system and device |
CN112052813A (en) * | 2020-09-15 | 2020-12-08 | 中国人民解放军军事科学院军事医学研究院 | Method, device, electronic device and readable storage medium for identifying translocation between chromosomes |
CN112288706A (en) * | 2020-10-27 | 2021-01-29 | 武汉大学 | Automatic chromosome karyotype analysis and abnormality detection method |
CN112487941A (en) * | 2020-11-26 | 2021-03-12 | 华南师范大学 | Method, system and storage medium for identifying chromosome cluster and chromosome instance |
CN112508889A (en) * | 2020-11-26 | 2021-03-16 | 郑州大学 | Chromosome karyotype analysis system |
CN115330603A (en) * | 2022-10-17 | 2022-11-11 | 湖南自兴智慧医疗科技有限公司 | Human chromosome image rectification method based on deep learning convolutional neural network |
CN115375682A (en) * | 2022-10-24 | 2022-11-22 | 湖南自兴智慧医疗科技有限公司 | Chromosome Roche translocation abnormality detection method, system and storage medium |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103558404A (en) * | 2013-11-08 | 2014-02-05 | 麦克奥迪(厦门)医疗诊断***有限公司 | Automatic cell DNA (deoxyribonucleic acid) detecting and rechecking method based on digital slides |
-
2018
- 2018-08-31 CN CN201811013795.4A patent/CN109242842A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103558404A (en) * | 2013-11-08 | 2014-02-05 | 麦克奥迪(厦门)医疗诊断***有限公司 | Automatic cell DNA (deoxyribonucleic acid) detecting and rechecking method based on digital slides |
Non-Patent Citations (1)
Title |
---|
谭伟: ""人类染色体核型分析***研究"", 《中国优秀硕士学位论文全文数据库-基础科学辑》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111209900B (en) * | 2020-04-20 | 2020-09-01 | 湖南自兴智慧医疗科技有限公司 | Image processing method for positioning centromere according to feature extraction algorithm |
CN111209900A (en) * | 2020-04-20 | 2020-05-29 | 湖南自兴智慧医疗科技有限公司 | Image processing method for positioning centromere according to feature extraction algorithm |
CN111986183A (en) * | 2020-08-25 | 2020-11-24 | 中国科学院长春光学精密机械与物理研究所 | Chromosome scattergram image automatic segmentation and identification system and device |
CN112052813A (en) * | 2020-09-15 | 2020-12-08 | 中国人民解放军军事科学院军事医学研究院 | Method, device, electronic device and readable storage medium for identifying translocation between chromosomes |
CN112052813B (en) * | 2020-09-15 | 2023-12-19 | 中国人民解放军军事科学院军事医学研究院 | Method and device for identifying translocation between chromosomes, electronic equipment and readable storage medium |
CN112288706B (en) * | 2020-10-27 | 2022-06-24 | 武汉大学 | Automatic chromosome karyotype analysis and abnormality detection method |
CN112288706A (en) * | 2020-10-27 | 2021-01-29 | 武汉大学 | Automatic chromosome karyotype analysis and abnormality detection method |
CN112508889A (en) * | 2020-11-26 | 2021-03-16 | 郑州大学 | Chromosome karyotype analysis system |
CN112508889B (en) * | 2020-11-26 | 2022-09-13 | 郑州大学 | Chromosome karyotype analysis system |
CN112487941B (en) * | 2020-11-26 | 2023-03-14 | 华南师范大学 | Method, system and storage medium for identifying chromosome cluster and chromosome instance |
CN112487941A (en) * | 2020-11-26 | 2021-03-12 | 华南师范大学 | Method, system and storage medium for identifying chromosome cluster and chromosome instance |
CN115330603A (en) * | 2022-10-17 | 2022-11-11 | 湖南自兴智慧医疗科技有限公司 | Human chromosome image rectification method based on deep learning convolutional neural network |
CN115375682A (en) * | 2022-10-24 | 2022-11-22 | 湖南自兴智慧医疗科技有限公司 | Chromosome Roche translocation abnormality detection method, system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242842A (en) | Human chromosomal analytical equipment, equipment and storage medium based on image recognition | |
Fan et al. | Automatic pavement crack detection based on structured prediction with the convolutional neural network | |
CN104217221A (en) | Method for detecting calligraphy and paintings based on textural features | |
CN106203496B (en) | Hydrographic curve extracting method based on machine learning | |
CN101604384A (en) | Individualized fingerprint identification method | |
CN113506239B (en) | Strip steel surface defect detection method based on cross-stage local network | |
CN111415352A (en) | Cancer metastasis panoramic pathological section analysis method based on deep cascade network | |
Majid et al. | An Integrated Deep Learning Framework for Fruits Diseases Classification. | |
CN111723819A (en) | Grading identification method for leaf veins of plant leaves | |
CN110415222A (en) | A kind of spinning cake side face defects recognition methods based on textural characteristics | |
CN104331892B (en) | Morphology-based neuron recognizing and analyzing method | |
CN110390312A (en) | Chromosome automatic classification method and classifier based on convolutional neural networks | |
CN112200789B (en) | Image recognition method and device, electronic equipment and storage medium | |
Li et al. | Incremental learning of infrared vehicle detection method based on SSD | |
CN115775226B (en) | Medical image classification method based on transducer | |
CN113516193B (en) | Image processing-based red date defect identification and classification method and device | |
CN110197114B (en) | Automatic identification method and device for single neuron axon synaptic junction in whole brain range | |
Sridevy et al. | Image pre-processing techniques utilized for the plant identification: A review | |
Li et al. | HEp-2 cells staining patterns classification via wavelet scattering network and random forest | |
Chakroun et al. | Hierarchical convolutional neural networks for leaf disease detection | |
Gunawan et al. | Classification of Japanese fagaceae wood based on microscopic image analysis | |
CN115375682B (en) | Chromosome Roche translocation abnormality detection method, system and storage medium | |
CN110070011A (en) | A kind of monitoring method and device applied to intelligence manufacture assembly line | |
Nabila et al. | A Good Performance of Convolutional Neural Network Based on AlexNet in Domestic Indonesian Car Types Classification | |
CN116168223B (en) | Multi-mode-based peanut leaf spot disease grade detection method |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |