CN115063411A - Chromosome abnormal region segmentation detection method and system - Google Patents
Chromosome abnormal region segmentation detection method and system Download PDFInfo
- Publication number
- CN115063411A CN115063411A CN202210932614.8A CN202210932614A CN115063411A CN 115063411 A CN115063411 A CN 115063411A CN 202210932614 A CN202210932614 A CN 202210932614A CN 115063411 A CN115063411 A CN 115063411A
- Authority
- CN
- China
- Prior art keywords
- chromosome
- abnormal
- region
- size
- region segmentation
- 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
Images
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- 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
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for detecting chromosome abnormal region segmentation, which comprises the steps of firstly, acquiring chromosome data, and marking the outline of each chromosome in the abnormal region; then, generating a mask binary image according to the contour to obtain a mask image label of the chromosome abnormal region; training an abnormal chromosome segmentation model according to the obtained image label; and finally, inputting a pair of homologous chromosomes to the trained model, namely outputting a mask image region of the chromosome abnormal region. The chromosome abnormal region segmentation detection method and the chromosome abnormal region segmentation detection system can solve chromosome early warning of possible chromosome abnormality, are not limited in abnormal chromosome type and abnormal chromosome region, and can meet most chromosome detection requirements.
Description
Technical Field
The invention relates to a chromosome abnormal region segmentation detection method and a system, belonging to the technology of intelligent identification of biological genetic resources.
Background
In the process of karyotyping, the detection of chromosomal abnormalities is an important part of the work, however, due to the wide variety, complex distribution, and poor sectioning and banding levels, chromosomal abnormalities are difficult to detect.
In the process of karyotyping, the following methods are mainly used as techniques for detecting abnormal chromosomes: one is the conventional chromosome classification method, such as t (9;22), chromosome 9 and 22 ectopy. In general, the normal chromosome 9 and the inverted chromosome 9 are divided into two types of data, and a classification model is used to train a two-class model, so as to finally obtain a classification model about the normality and abnormality of the chromosome 9. Then, the same operation is performed on chromosome 22, and classification training is performed on normal chromosome 22 and abnormal chromosome 22. Finally, inputting a single chromosome 9 and a single chromosome 22 to a No. 9 binary model and a No. 22 binary model respectively. And obtaining whether the chromosome is abnormal or not according to whether the output is normal or not.
The method is poor in applicability, and classification model training is required for each chromosome. Human beings have 24 types of chromosomes, meaning that 24 classification models are required, computational resources and costs are increased, and speed is extremely slow. And the method only can know that the chromosome number has possible abnormality but cannot be accurate to the strip region. Therefore, the simple and effective abnormal chromosome detection method has great significance for chromosome karyotype analysis.
Disclosure of Invention
The technical problem solved by the invention is as follows: aiming at the problem of complex calculation of chromosome nucleation image abnormality detection by a conventional chromosome classification method, a chromosome abnormal region segmentation detection method and a system are provided.
The invention is realized by adopting the following technical scheme:
firstly, the invention provides a chromosome abnormal region segmentation detection method, which comprises the following steps:
s1, collecting chromosome data, and marking the abnormal region outline of each chromosome;
s2, generating a mask binary image according to the contour to obtain a mask image label of the chromosome abnormal region;
s3, training an abnormal chromosome segmentation model;
and S4, inputting a pair of homologous chromosomes into the trained model, namely outputting a mask image region of the chromosome abnormality region.
In the method for detecting a chromosomal abnormality region according to the present invention, the step S1 further includes the steps of:
s11, cutting the chromosome and the homologous chromosome thereof from the metaphase map, removing the background outside the chromosome through the outer contour of the chromosome, and resize to a uniform size;
and S12, marking the bands of the chromosome abnormal region to obtain a mask of the chromosome abnormal region.
In the method for detecting chromosome abnormality region segmentation of the present invention, further, in step S2, the two homologous chromosomes are aligned and then spliced into a dual-channel matrix, and the dual-channel matrix is input to a convolutional neural network, so as to obtain a striped mask with an obvious difference between the two homologous chromosomes.
In the method for detecting chromosome abnormal region segmentation of the present invention, further, the convolutional neural network performs feature extraction and feature coding on an input chromosome image, and converts plane information into channel information, and a specific convolution calculation formula is as follows:
wherein W is the size of the input layer, K is the size of the convolution kernel, P is the filled pixel, S is the convolution kernel sliding step length, and O is the size of the feature map after convolution.
In the method for detecting a chromosomal abnormality region according to the present invention, the step S3 further includes the steps of:
s31, after the convolutional neural network coding is completed, performing up-sampling on the feature map obtained in the step S2 by using a deconvolution operation, and effectively restoring the plane and channel information of the feature map;
s32, obtaining a final feature map through up-sampling, and supervising the final feature map to enable the final feature map to be capable of learning according to a specified direction, enhancing the constraint of the chromosome shape and reducing the outline enlargement and deletion of the chromosome in the segmentation process;
s33, monitoring pixels in the chromosome, enabling the chromosome and the noise background to be accurately segmented, and improving the capability of the model for distinguishing impurities from chromosomes.
In the method for detecting chromosome abnormality region segmentation according to the present invention, further, in step S31, the formula adopted by the upsampling is as follows:
wherein, O1 is the size of the feature map after deconvolution in the upsampling process, S1 is the sliding step size of the upsampling convolution kernel, W1 is the size of the input layer feature map in the upsampling process, K1 is the size of the upsampling convolution kernel, and P1 is the pixel filled with the upsampling.
In the method for detecting chromosome abnormality region segmentation of the present invention, further, in step S32, a dice loss function is used to supervise a connected domain region of a final feature map, so that two homologous chromosome abnormality regions can be better and completely segmented, and a formula specifically adopted is as follows:
wherein, Dice is the overlapping degree between the label and the prediction connected domain, Pred is a prediction value, namely a model output mask image, and True is a True value, namely a mask image generated by labeling.
In the method for detecting chromosome abnormal region segmentation of the present invention, further, in step S33, chromosome pixel supervision is performed by using the following formula:
where BCE is the binary cross entropy between the label pixel value and the predicted pixel value, y is the label 0 or 1, a is the output of the convolutional neural network through the sigmoid function, the range is (0, 1), and n is the number of samples.
In the method for detecting a chromosome abnormality region by segmentation of the present invention, further, in step S4, the input homologous chromosomes are two-channel images formed by splicing homologous chromosomes.
The invention also provides a chromosome abnormal region segmentation detection system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the method when executing the computer program.
According to the method, the image characteristics of the abnormal region in the chromosome metaphase map are extracted by adopting the stacked convolution, the characteristic map is compressed by pooling, and the compressed image is restored by up-sampling and convolution, so that the method realizes a brand-new mode for carrying out segmentation detection on the abnormal striae interval of the chromosome, abandons the traditional classification method, avoids the problems of low speed and wide computing resource consumption, and can find almost all abnormal chromosomes by only one neural network model. The method has the advantages of high detection speed and wide application range, and can detect abnormal chromosomes and mark specific areas of chromosome abnormality.
In summary, the chromosome abnormal region segmentation detection method and system provided by the invention can solve chromosome early warning that a chromosome is possibly abnormal, do not limit abnormal chromosome types and abnormal chromosome regions, and can meet most chromosome detection requirements.
The invention is further described with reference to the following figures and detailed description.
Drawings
FIG. 1 is a schematic flow chart of a method for detecting chromosome abnormality region segmentation according to the present invention.
FIG. 2 is an exemplary diagram of the input model before training after the annotation is completed.
Fig. 3 is a schematic diagram of chromosome image labeling and prediction results in an example.
Detailed Description
To further illustrate the technical means and effects of the present invention, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced device or element must have a particular orientation, be constructed in a particular orientation, and be operated, and thus should not be considered as limiting the present application. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
In this application, the word "exemplary" is used to mean "serving as an example, instance, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for the purpose of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes are not set forth in detail in order to avoid obscuring the description of the present application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
As shown in fig. 1, the method for detecting chromosome abnormality region segmentation provided by the present invention comprises the following steps:
and step S1, acquiring chromosome data, and marking the abnormal region contour of each chromosome.
The chromosome data is selected from metaphase maps of chromosomes and their homologues. The method comprises the following specific steps:
substep S11, cropping the chromosome and its homologous chromosomes from the metaphase map, removing the background outside the chromosome by the outline of the chromosome, resize to uniform size, and generally setting the picture pixels to 128 × 128 wide and high resolution.
And a substep S12 of marking the belt of the chromosome abnormal region in the cut chromosome picture to obtain a mask of the chromosome abnormal region. The chromosome karyotype analyst marks the outline of the abnormal region, and the marking of the chromosome abnormal region with the stripe can be carried out by a marking tool labelme.
And step S2, generating a mask binary image according to the contour to obtain a mask image label of the chromosome abnormal region.
The specific process is as follows: straightening and straightening the two homologous chromosomes, splicing the two homologous chromosomes into a double-channel matrix, and inputting the double-channel matrix into a convolutional neural network to obtain the striped mask with obvious difference between the two homologous chromosomes. The convolutional neural network performs feature extraction and feature coding on the input chromosome image, and converts plane information into channel information, and a specific convolution calculation formula is shown as formula (1).
In equation (1), W is the size of the input layer, K is the size of the convolution kernel, P is the filled pixel, S is the convolution kernel sliding step, and O is the feature size after convolution.
And step S3, training an abnormal chromosome segmentation model.
The training process of the segmentation model comprises the following sub-steps:
in the substep S31, after encoding by convolutional neural network is completed, the feature map obtained in the step S2 is up-sampled by using a deconvolution operation, and the up-sampling formula is as shown in formula (2).
In equation (2), where O1 is the feature size after deconvolution by the upsampling process, S1 is the upsampling convolution kernel sliding step size, W1 is the size of the upsampling process input layer feature, K1 is the upsampling convolution kernel size, and P1 is the upsampled filled pixel.
And a substep S32 of obtaining a final feature map through up-sampling, monitoring the final feature map to enable the final feature map to learn according to a specified direction, wherein a dice loss function is used for monitoring a connected domain area of the final feature map to enable two homologous chromosome abnormal areas to be well and completely segmented, and a formula specifically adopted is a formula (3).
Wherein, Dice is the overlapping degree between the label and the prediction connected domain, Pred is a prediction value, namely a model output mask image, and True is a True value, namely a mask image generated by labeling.
And a substep S33 of monitoring pixels in the chromosome so as to correctly segment the chromosome and the noise background, wherein the process adopts a formula (4) to monitor the chromosome pixels.
Where BCE is the binary cross entropy between the label pixel value and the predicted pixel value, y is the label 0 or 1, a is the output of the convolutional neural network through the sigmoid function, the range is (0, 1), and n is the number of samples.
And S4, detecting the abnormal chromosome region after the model training is finished, inputting a two-channel image formed by splicing a pair of homologous chromosomes into the trained model, and outputting the mask image region of the abnormal chromosome region.
The invention also provides a chromosome abnormal region segmentation detection system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the method when executing the computer program.
As shown in fig. 2 and fig. 3, fig. 2 is a two-channel image obtained by stitching two homologous chromosomes used for model training. And (5) uniformly sizing the spliced pictures to 128 × 128, and inputting the sized pictures into the Unet model. And obtaining an output two-channel mask graph with 128 × 128 size, calculating the two-channel mask graph with the two binary graphs in the graph 2, adding a connected domain overlapping degree loss Dice loss and a binary cross entropy BCE loss between pixels to obtain an overall loss, and performing back propagation to train the abnormal chromosome segmentation model in a reciprocating way.
And after a trained abnormal chromosome segmentation model is obtained, outputting a dual-channel homologous chromosome image to obtain an area mask image of the abnormality between the two homologous chromosomes, and drawing the outline on the original chromosome image, wherein the area is the area marked in the frame in the image 3 and is the section corresponding to the chromosome abnormality.
The above description is only an embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the protection of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention are still within the protection scope of the technical solution of the present invention.
Claims (10)
1. A method for detecting chromosome abnormality region segmentation is characterized by comprising the following steps:
s1, collecting chromosome data, and marking the abnormal region outline of each chromosome;
s2, generating a mask binary image according to the contour to obtain a mask image label of the chromosome abnormal region;
s3, training an abnormal chromosome segmentation model;
and S4, inputting a pair of homologous chromosomes to the trained model, namely outputting a mask image region of the chromosome abnormal region.
2. The chromosome abnormality region segmentation detection method according to claim 1, characterized in that: the step S1 includes the following steps:
s11, cutting the chromosome and the homologous chromosome thereof from the metaphase map, removing the background outside the chromosome through the outer contour of the chromosome, and resize to a uniform size;
and S12, marking the bands of the chromosome abnormal region to obtain a mask of the chromosome abnormal region.
3. The method for detecting chromosomal abnormality region according to claim 1, wherein: in step S2, the two homologous chromosomes are straightened and aligned, and then spliced into a two-channel matrix, and input to a convolutional neural network, so as to obtain a striped mask with an obvious difference between the two homologous chromosomes.
4. The method for detecting chromosomal abnormality region segmentation according to claim 3, characterized in that: the convolutional neural network performs feature extraction and feature coding on an input chromosome image, and converts plane information into channel information, wherein a specific convolution calculation formula is as follows:
wherein W is the size of the input layer, K is the size of the convolution kernel, P is the filled pixel, S is the convolution kernel sliding step length, and O is the size of the feature map after convolution.
5. The method for detecting chromosomal abnormality region segmentation according to claim 3, characterized in that: the step S3 includes the following steps:
s31, after the convolutional neural network coding is completed, performing up-sampling on the characteristic diagram obtained in the step S2 by using a deconvolution operation;
s32, obtaining a final feature map through up-sampling, and supervising the final feature map to enable the final feature map to learn according to a specified direction;
and S33, monitoring pixels in the chromosome so that the chromosome and the noise background can be accurately segmented.
6. The method for detecting chromosomal abnormality region segmentation according to claim 5, wherein: in step S31, the formula used for the upsampling is as follows:
wherein O1 is the size of the feature map after deconvolution in the upsampling process, S1 is the sliding step size of the upsampling convolution kernel, W1 is the size of the input layer feature map in the upsampling process, K1 is the size of the upsampling convolution kernel, and P1 is the pixel filled with the upsampling.
7. The method for detecting chromosomal abnormality region segmentation according to claim 5, wherein: in step S32, a dice loss function is used to supervise the connected domain region of the final feature map, so that two homologous chromosome abnormality regions can be better and completely segmented, and the formula specifically adopted is as follows:
wherein, Dice is the overlapping degree between the label and the prediction connected domain, Pred is a prediction value, namely a model output mask image, and True is a True value, namely a mask image generated by labeling.
8. The method for detecting chromosomal abnormality region segmentation according to claim 5, wherein: in step S33, chromosome pixel supervision is performed by using the following formula:
where BCE is the binary cross entropy between the label pixel value and the predicted pixel value, y is the label 0 or 1, a is the output of the convolutional neural network through the sigmoid function, the range is (0, 1), and n is the number of samples.
9. The method for detecting chromosomal abnormality region according to claim 1, wherein: in step S4, the input homologous chromosomes are two-channel maps spliced by homologous chromosomes.
10. A chromosome abnormality region segmentation detection system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: the processor, when executing the computer program, implements the method of any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210932614.8A CN115063411A (en) | 2022-08-04 | 2022-08-04 | Chromosome abnormal region segmentation detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210932614.8A CN115063411A (en) | 2022-08-04 | 2022-08-04 | Chromosome abnormal region segmentation detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115063411A true CN115063411A (en) | 2022-09-16 |
Family
ID=83207817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210932614.8A Pending CN115063411A (en) | 2022-08-04 | 2022-08-04 | Chromosome abnormal region segmentation detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115063411A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111402259A (en) * | 2020-03-23 | 2020-07-10 | 杭州健培科技有限公司 | Brain tumor segmentation method based on multi-level structure relation learning network |
| CN111951266A (en) * | 2020-09-01 | 2020-11-17 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligence recognition analysis method for chromosome aberration |
| US20200410668A1 (en) * | 2019-06-26 | 2020-12-31 | Ever Fortune.Ai Co., Ltd. | Method for training a separation model for separation of overlapping chromosome based on simulation, and method and system for implementing separation of overlapping chromosome using the separation model |
| CN112288706A (en) * | 2020-10-27 | 2021-01-29 | 武汉大学 | Automatic chromosome karyotype analysis and abnormality detection method |
| CN112711983A (en) * | 2020-12-08 | 2021-04-27 | 湖南自兴智慧医疗科技有限公司 | Nuclear pattern analysis system, method, electronic device and readable storage medium |
| CN112819914A (en) * | 2021-02-05 | 2021-05-18 | 北京航空航天大学 | PET image processing method |
| CN112949550A (en) * | 2021-03-19 | 2021-06-11 | 中国科学院空天信息创新研究院 | Water body identification method, system and medium based on deep learning |
| CN113658150A (en) * | 2021-08-23 | 2021-11-16 | 西安交通大学 | Chromosome automatic segmentation and classification method based on deep learning |
| CN114842472A (en) * | 2022-07-04 | 2022-08-02 | 杭州德适生物科技有限公司 | Method and device for detecting chromosome structure abnormality based on deep learning |
| CN114841294A (en) * | 2022-07-04 | 2022-08-02 | 杭州德适生物科技有限公司 | Classifier model training method and device for detecting chromosome structure abnormality |
-
2022
- 2022-08-04 CN CN202210932614.8A patent/CN115063411A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200410668A1 (en) * | 2019-06-26 | 2020-12-31 | Ever Fortune.Ai Co., Ltd. | Method for training a separation model for separation of overlapping chromosome based on simulation, and method and system for implementing separation of overlapping chromosome using the separation model |
| CN111402259A (en) * | 2020-03-23 | 2020-07-10 | 杭州健培科技有限公司 | Brain tumor segmentation method based on multi-level structure relation learning network |
| CN111951266A (en) * | 2020-09-01 | 2020-11-17 | 厦门汉舒捷医疗科技有限公司 | Artificial intelligence recognition analysis method for chromosome aberration |
| CN112288706A (en) * | 2020-10-27 | 2021-01-29 | 武汉大学 | Automatic chromosome karyotype analysis and abnormality detection method |
| CN112711983A (en) * | 2020-12-08 | 2021-04-27 | 湖南自兴智慧医疗科技有限公司 | Nuclear pattern analysis system, method, electronic device and readable storage medium |
| CN112819914A (en) * | 2021-02-05 | 2021-05-18 | 北京航空航天大学 | PET image processing method |
| CN112949550A (en) * | 2021-03-19 | 2021-06-11 | 中国科学院空天信息创新研究院 | Water body identification method, system and medium based on deep learning |
| CN113658150A (en) * | 2021-08-23 | 2021-11-16 | 西安交通大学 | Chromosome automatic segmentation and classification method based on deep learning |
| CN114842472A (en) * | 2022-07-04 | 2022-08-02 | 杭州德适生物科技有限公司 | Method and device for detecting chromosome structure abnormality based on deep learning |
| CN114841294A (en) * | 2022-07-04 | 2022-08-02 | 杭州德适生物科技有限公司 | Classifier model training method and device for detecting chromosome structure abnormality |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
|---|---|---|
| CN111241947B (en) | Training method and device for target detection model, storage medium and computer equipment | |
| CN112017189B (en) | Image segmentation method and device, computer equipment and storage medium | |
| CN109493346A (en) | It is a kind of based on the gastric cancer pathology sectioning image dividing method more lost and device | |
| WO2022237139A1 (en) | Lanesegnet-based lane line detection method and system | |
| US20140016843A1 (en) | Computer-assisted karyotyping | |
| CN110674704A (en) | Crowd density estimation method and device based on multi-scale expansion convolutional network | |
| CN114299066B (en) | Defect detection method and device based on salient feature pre-extraction and image segmentation | |
| CN111126175A (en) | Facial image recognition algorithm based on deep convolutional neural network | |
| US20200160547A1 (en) | Method of predicting depth values of lines, method of outputting three-dimensional (3d) lines, and apparatus thereof | |
| CN114612472B (en) | SegNet improvement-based leather defect segmentation network algorithm | |
| CN112419202B (en) | Automatic wild animal image recognition system based on big data and deep learning | |
| WO2021164280A1 (en) | Three-dimensional edge detection method and apparatus, storage medium and computer device | |
| CN111899253B (en) | Abnormal judgment analysis method and device for fetal craniocerebral section image | |
| CN115063411A (en) | Chromosome abnormal region segmentation detection method and system | |
| CN113066080A (en) | Method and device for identifying slice tissue, cell identification model and tissue segmentation model | |
| CN117152484B (en) | Small target cloth flaw detection method based on improved YOLOv5s | |
| CN110276759A (en) | A kind of bad line defect diagnostic method of Mobile phone screen based on machine vision | |
| CN113034514A (en) | Sky region segmentation method and device, computer equipment and storage medium | |
| CN114511523B (en) | Gastric cancer molecular subtype classification method and device based on self-supervision learning | |
| CN116597343A (en) | Expressway weather identification method and device based on ensemble learning algorithm | |
| CN114972202A (en) | Ki67 pathological cell rapid detection and counting method based on lightweight neural network | |
| CN113963008A (en) | Small target semantic segmentation method and system for high-resolution image | |
| CN116137061B (en) | Training method and device for quantity statistical model, electronic equipment and storage medium | |
| CN112949500A (en) | Improved YOLOv3 lane line detection method based on spatial feature coding | |
| CN111127503A (en) | Method, device and storage medium for detecting the pattern of a vehicle tyre |
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: 20220916 |