CN115050024A - Interpretable granulocyte intelligent real-time identification method and system - Google Patents
Interpretable granulocyte intelligent real-time identification method and system Download PDFInfo
- Publication number
- CN115050024A CN115050024A CN202210706081.1A CN202210706081A CN115050024A CN 115050024 A CN115050024 A CN 115050024A CN 202210706081 A CN202210706081 A CN 202210706081A CN 115050024 A CN115050024 A CN 115050024A
- Authority
- CN
- China
- Prior art keywords
- blood
- staining
- image
- prior information
- interpretable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/693—Acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses an interpretable granulocyte intelligent real-time identification method and an interpretable granulocyte intelligent real-time identification system. The method comprises the following steps of collecting a blood staining slide image, and acquiring blood cell staining prior information based on the blood staining slide image; obtaining a detection target cell based on a target cell detection instruction and the blood cell staining prior information; constructing and training a deep learning model; and based on the trained deep learning model, completing intelligent identification and prediction of the category of the granulocyte panoramic microscopic image. Respectively training and predicting through a plurality of convolutional neural network models, and preferentially selecting; image information is adopted by adopting an RGB conversion method, so that the image information is not easily influenced by image noise; and expanding according to the position of the target pixel to improve the accuracy of identification prediction.
Description
Technical Field
The invention belongs to the technical field of granulocyte detection, and particularly relates to an interpretable intelligent real-time granulocyte identification method and an interpretable intelligent real-time granulocyte identification system.
Background
White Blood Cells (WBCs) are an important class of immune cells in human body fluids, an important component of the immune system, and can be classified by morphological differences into the presence and absence of particles. The class of particles, which we refer to as granulocytes, is divided into neutrophils, eosinophils and basophils, according to the staining properties of the particles in the cytoplasm, with the proportion of neutrophils being the greatest; one class that is particle-free is monocytes and lymphocytes. Granulocytes are differentiated from hematopoietic stem cells in the bone marrow, a process that has six stages: primitive granulocytes, promyelocytes, mesogranulocytes, metagranulocytes, rod-shaped granulocytes, and lobulated granulocytes. Since specific granules, i.e., S granules, appear in the neutrophils, eosinophils, and basophils, the S granules are classified into neutrophils, eosinophils, and basophils according to the granule types, and so on.
The application of medical image processing techniques and pattern recognition methods to check the white blood cell recognition classification count is one of the important means for medical assisted diagnosis. Because of the abundance and diversity of granulocytes, there is a need to further classify and identify the morphological features and number of granulocytes of various types to help further disease diagnosis. However, due to the high cost of medical image generation and the need for privacy protection, the typical data set obtained in practice is small, which emphasizes the importance of image preprocessing such as data enhancement, regularization and cross validation. To acquire more data, minor changes to the existing data set, such as rotation, flipping, shifting, etc., are required.
Disclosure of Invention
In order to solve the technical problems, the invention provides an interpretable granulocyte intelligent real-time identification method and an interpretable granulocyte intelligent real-time identification system, wherein a plurality of convolutional neural network models are respectively trained and predicted, and are preferentially selected; image information is adopted by adopting an RGB conversion method, so that the influence of image noise is not easy to influence; and expanding according to the position of the target pixel to improve the accuracy of identification prediction.
In order to achieve the above object, the present invention provides an interpretable granulocyte intelligent real-time identification method, which comprises the following steps:
collecting a blood staining slide image, and acquiring blood cell staining prior information based on the blood staining slide image;
obtaining a detection target cell based on a target cell detection instruction and the blood cell staining prior information;
constructing and training a deep learning model;
and based on the trained deep learning model, completing intelligent identification and prediction of the category of the panoramic microscopic image of the detection target cell.
Optionally, obtaining the prior information on blood cell staining comprises:
performing three-channel separation on the blood staining slide image;
and performing machine learning on the blood staining slide image separated by the three channels to obtain blood cell staining prior information.
Optionally, the blood cell staining prior information comprises: red blood cell prior information, white blood cell prior information, and platelet prior information.
Optionally, the method for obtaining the detection target cell comprises:
acquiring a target cell detection instruction, and acquiring a corresponding target pixel position based on the target cell detection instruction;
performing image cropping on the blood-stained slide image based on the target pixel location;
performing RGB three-channel separation on the cut blood staining slide image;
and matching the separated blood staining slide image with the blood cell staining prior information to obtain a detection target cell.
Optionally, the method for obtaining the corresponding target pixel position is:
and scanning the blood staining slide image by adopting an OpenCV image reading and channel separation method based on the target cell detection instruction to obtain a corresponding target pixel position.
Optionally, the method of image cropping for the blood stained slide image is as follows:
based on the target pixel location, the blood-stained slide image is cropped to a square of 360 pixels by 360 pixels centered on the pixel.
Optionally, the method for training the deep learning model includes:
acquiring a data set based on a granulocyte panoramic microscopic image, wherein the data set comprises a training set and a test set;
training the deep learning model based on the training set;
and testing the trained deep learning model based on a test set.
On the other hand, in order to achieve the above object, the invention provides an interpretable granulocyte intelligent real-time recognition system, which comprises a first acquisition module, a second acquisition module, a construction module and a classification module;
the first acquisition module is used for acquiring a blood staining slide image and acquiring blood cell staining prior information based on the blood staining slide image;
the second acquisition module is used for acquiring a detection target cell based on a target cell detection instruction and the blood cell staining prior information;
the building module is used for building and training a deep learning model;
the classification module is used for completing intelligent identification and prediction of the category of the panoramic microscopic image of the detection target cell based on the trained deep learning model.
Optionally, obtaining the prior information on blood cell staining comprises:
performing three-channel separation on the blood staining slide image;
and performing machine learning on the blood staining slide image separated by the three channels to obtain blood cell staining prior information.
Optionally, the blood cell staining prior information comprises: red blood cell prior information, white blood cell prior information, and platelet prior information.
Compared with the prior art, the invention has the following advantages and technical effects:
the method learns the blood stained slide image by a machine learning method to obtain the prior information of the stained slide image of the red blood cells, the white blood cells and the platelets; the image information is adopted by adopting an RGB conversion method, so that the image information is not easily influenced by image noise, the position of the target cell is automatically and quickly acquired in an all-round way, the gradual searching is reduced, and the calculated amount is reduced; the plurality of convolutional neural network models are respectively trained and predicted, preferentially selected, expanded according to the target pixel position to improve the accuracy and efficiency of recognition and prediction, and automatically give a prompt when a low probability type value appears.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, are included to provide a further understanding of the application, and the description of the exemplary embodiments of the application are intended to be illustrative of the application and are not intended to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of an interpretable method for identifying granulocytes in real time according to an embodiment of the present invention;
fig. 2 is a schematic diagram of image cropping according to a location target pixel position according to a first embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than here.
Example one
As shown in FIG. 1, the present invention provides an interpretable granulocyte intelligent real-time identification method, comprising the following steps:
collecting a blood staining slide image, and acquiring blood cell staining prior information based on the blood staining slide image:
further, a slide is made by collecting blood for staining, and an image of the entire slide is collected.
And carrying out three-channel separation on the image, learning by a machine learning method, acquiring prior information of red blood cells, white blood cells and platelets in the blood cell stained slide, and setting a terminal prior information input window.
Based on the target cell detection instruction and the prior blood cell staining information, obtaining a detection target cell:
further, first, the prepared blood-stained slide is set in the microsystem apparatus, and when the detection of a target cell to be detected, for example, a granulocyte, is input by inputting the "manual target cell input terminal", the leukocyte number is input.
After the system obtains a target cell detection instruction, the system scans the slide panorama through methods of image reading, channel separation and the like of OpenCV to obtain a corresponding target position.
And performing image cropping according to the position of the positioning target pixel and a square with the pixel as the center and the size of 360 pixels by 360 pixels, as shown in fig. 2.
And performing RGB three-channel separation according to the cut image, and verifying whether the cell is a target cell or not again through the prior information. That is, the range of the RGB channel values of the target cell image is divided in advance, if the clipped image is traversed by each pixel, if the RGB channel values all fall within the prior range, the target cell is determined.
Constructing and training a deep learning model;
furthermore, a large-magnification image of 8 kinds of granulocytes such as primitive granulocytes, promyelocytes, mesogranulocytes, metagranulocytes, rod-shaped granulocytes, lobular granulocytes, eosinophils, basophils, etc., which are microscopically obtained from a granulocytic overall microscopic image, is collected.
And dividing a training set and a testing set (ratio 8: 2) to perform model training. For example, VGG11, VGG13, VGG16, VGG19, Resnet34, Resnet50, Resnet101, Resnet152, Densenet121, Densenet169, Densenet201, Densenet264, Moblenet and IncepotionResNet models are trained, and the model with the best accuracy is obtained as the classifier.
Firstly, the hyper-parameters are optimized to obtain the optimal momentum factor, learning rate, batch size, regularization method selection, feature map size and other parameters. The model is then trained according to the parameter values.
Based on the trained deep learning model, completing intelligent identification and prediction of the category of the whole microscopic image of the detection target cell:
further, after validation, the images are sent to a trained model classifier for prediction. I.e. the highest accuracy model of the above-mentioned training models.
In model prediction, the result of the model output is an 8-dimensional vector, e.g., (p) 1 ,p 2 ,p 3 ,p 4 ,p 5 ,p 6 ,p 7 ,p 8 ) Wherein the value is a probability value. And if the maximum value of the eight values of the output vector is less than 0.7, classifying the result value into other classes, giving an early warning and prompting an inspector to manually check.
Example two
The invention provides an interpretable granulocyte intelligent real-time recognition system, which comprises,
the system comprises a first acquisition module, a second acquisition module, a construction module and a classification module;
the first acquisition module is used for acquiring a blood staining slide image, and acquiring blood cell staining prior information based on the blood staining slide image:
further, a slide glass is prepared by collecting blood for staining, and an image of the entire slide glass is collected.
And (3) carrying out three-channel separation on the image, learning by a machine learning method, acquiring prior information of red blood cells, white blood cells and platelets in the blood cell stained slide, and setting a terminal prior information input window.
The second acquisition module is used for acquiring a detection target cell based on the target cell detection instruction and the blood cell staining prior information:
further, the prepared blood-stained slide is first placed in a microsystem apparatus, and a leukocyte number is input when the detection of a detection target cell, for example, a granulocyte, is input by inputting a "manual target cell input terminal".
After the system obtains a target cell detection instruction, the system scans the slide panorama through methods of image reading, channel separation and the like of OpenCV to obtain a corresponding target position.
And performing image cropping according to the position of the positioning target pixel and a square with the pixel as the center and the size of 360 pixels by 360 pixels, as shown in fig. 2.
And performing RGB three-channel separation according to the cut image, and verifying whether the cell is a target cell or not again through the prior information. That is, the range of the RGB channel values of the target cell image is divided in advance, if the clipped image is traversed by each pixel, if the RGB channel values all fall within the prior range, the target cell is determined.
The building module is used for building and training a deep learning model:
furthermore, a large-magnification image of 8 kinds of granulocytes such as primitive granulocytes, promyelocytes, mesogranulocytes, metagranulocytes, rod-shaped granulocytes, lobular granulocytes, eosinophils, basophils, etc., which are microscopically obtained from a granulocytic overall microscopic image, is collected.
The training set and the test set are divided (ratio 8: 2) and model training is performed. For example, VGG11, VGG13, VGG16, VGG19, Resnet34, Resnet50, Resnet101, Resnet152, Densenet121, Densenet169, Densenet201, Densenet264, Mobleenet and IncepotionResNet models are trained, and the model with the best precision is obtained as the classifier.
Firstly, the hyper-parameters are optimized to obtain the optimal momentum factor, learning rate, batch size, regularization method selection, feature map size and other parameters. The model is then trained according to the parameter values.
The classification module is used for completing intelligent identification and prediction of the category of the panoramic microscopic image of the detected target cell based on the trained deep learning model:
further, after validation, the images are sent to a trained model classifier for prediction. I.e. the highest accuracy model in the above-mentioned training models.
In model prediction, the result of the model output is an 8-dimensional vector, e.g., (p) 1 ,p 2 ,p 3 ,p 4 ,p 5 ,p 6 ,p 7 ,p 8 ) Wherein the value is a probability value. And if the maximum value of the eight values of the output vector is less than 0.7, classifying the result value into other classes, giving an early warning and prompting an inspector to manually check.
The invention provides an interpretable intelligent real-time granulocyte identification method and an interpretable intelligent real-time granulocyte identification system, which are used for intelligently identifying and predicting types of 8-granulocyte full-scope microscopic images such as primitive granulocytes, promyelocytes, mesogranulocytes, promyelocytes, rod-shaped granulocytes, lobular granulocytes, eosinophils, basophils and the like. In the proposed model, in order to automatically classify the granulocyte images subjected to data enhancement preprocessing into 8 cell classes, convolutional neural network models with different architectures, including VGG16, ResNet, densnet and the like, are used first, and then the granulocyte panoramic microscopic images are pre-trained and predicted by using the architectures respectively.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An interpretable granulocyte intelligent real-time identification method is characterized by comprising the following steps,
collecting a blood staining slide image, and acquiring blood cell staining prior information based on the blood staining slide image;
obtaining a detection target cell based on a target cell detection instruction and the blood cell staining prior information;
constructing and training a deep learning model;
and based on the trained deep learning model, completing intelligent identification and prediction of the category of the panoramic microscopic image of the detection target cell.
2. The intelligent real-time interpretable granulocyte identification method of claim 1, wherein obtaining a priori information about blood cell staining comprises:
performing three-channel separation on the blood staining slide image;
and performing machine learning on the blood staining slide image separated by the three channels to obtain blood cell staining prior information.
3. The intelligent real-time interpretable granulocyte identification method of claim 2, wherein the blood cell staining prior information comprises: red blood cell prior information, white blood cell prior information, and platelet prior information.
4. The method for intelligently and real-time identifying interpretable granulocytes of claim 1, wherein the method for obtaining the target cells for detection comprises:
acquiring a target cell detection instruction, and acquiring a corresponding target pixel position based on the target cell detection instruction;
performing image cropping on the blood-stained slide image based on the target pixel location;
performing RGB three-channel separation on the cut blood staining slide image;
and matching the separated blood staining slide image with the blood cell staining prior information to obtain a detection target cell.
5. The method for intelligent real-time identification of interpretable granulocytes of claim 4, wherein the method for obtaining the corresponding target pixel location comprises:
and scanning the blood staining slide image by adopting an OpenCV image reading and channel separation method based on the target cell detection instruction to obtain a corresponding target pixel position.
6. The method for intelligent real-time identification of interpretable granulocytes of claim 4, wherein the method for image cropping of the blood-stained slide image comprises:
based on the target pixel location, the blood-stained slide image is cropped to a square of 360 pixels by 360 pixels centered on the pixel.
7. The method for intelligent real-time identification of interpretable granulocytes of claim 1, wherein the method for training the deep learning model comprises:
acquiring a data set based on a granulocyte panoramic microscopic image, wherein the data set comprises a training set and a test set;
training the deep learning model based on the training set;
and testing the trained deep learning model based on a test set.
8. An interpretable granulocyte intelligent real-time recognition system is characterized by comprising,
the system comprises a first acquisition module, a second acquisition module, a construction module and a classification module;
the first acquisition module is used for acquiring a blood staining slide image and acquiring blood cell staining prior information based on the blood staining slide image;
the second acquisition module is used for acquiring a detection target cell based on a target cell detection instruction and the blood cell staining prior information;
the building module is used for building and training a deep learning model;
the classification module is used for completing intelligent identification and prediction of the category of the panoramic microscopic image of the detection target cell based on the trained deep learning model.
9. The system of claim 8, wherein obtaining a priori information on staining of blood cells comprises:
performing three-channel separation on the blood staining slide image;
and performing machine learning on the blood staining slide image separated by the three channels to obtain blood cell staining prior information.
10. The system of claim 9, wherein the blood cell staining prior information comprises: red blood cell prior information, white blood cell prior information, and platelet prior information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210706081.1A CN115050024B (en) | 2022-06-21 | 2022-06-21 | Intelligent real-time identification method and system for granulocytes with interpretability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210706081.1A CN115050024B (en) | 2022-06-21 | 2022-06-21 | Intelligent real-time identification method and system for granulocytes with interpretability |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115050024A true CN115050024A (en) | 2022-09-13 |
| CN115050024B CN115050024B (en) | 2023-08-22 |
Family
ID=83164014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210706081.1A Active CN115050024B (en) | 2022-06-21 | 2022-06-21 | Intelligent real-time identification method and system for granulocytes with interpretability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115050024B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017519985A (en) * | 2014-06-16 | 2017-07-20 | シーメンス・ヘルスケア・ダイアグノスティックス・インコーポレーテッドＳｉｅｍｅｎｓ Ｈｅａｌｔｈｃａｒｅ Ｄｉａｇｎｏｓｔｉｃｓ Ｉｎｃ． | Digital holographic microscopy data analysis for hematology |
| CN107977682A (en) * | 2017-12-19 | 2018-05-01 | 南京大学 | Lymph class cell sorting method and its device based on the enhancing of polar coordinate transform data |
| CN109063766A (en) * | 2018-07-31 | 2018-12-21 | 湘潭大学 | A kind of image classification method based on identification prediction sparse decomposition model |
| CN112597852A (en) * | 2020-12-15 | 2021-04-02 | 深圳大学 | Cell sorting method, cell sorting device, electronic apparatus, and storage medium |
| CN113869124A (en) * | 2021-08-30 | 2021-12-31 | 桂林优利特医疗电子有限公司 | Deep learning-based blood cell morphology classification method and system |
-
2022
- 2022-06-21 CN CN202210706081.1A patent/CN115050024B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017519985A (en) * | 2014-06-16 | 2017-07-20 | シーメンス・ヘルスケア・ダイアグノスティックス・インコーポレーテッドＳｉｅｍｅｎｓ Ｈｅａｌｔｈｃａｒｅ Ｄｉａｇｎｏｓｔｉｃｓ Ｉｎｃ． | Digital holographic microscopy data analysis for hematology |
| CN107977682A (en) * | 2017-12-19 | 2018-05-01 | 南京大学 | Lymph class cell sorting method and its device based on the enhancing of polar coordinate transform data |
| CN109063766A (en) * | 2018-07-31 | 2018-12-21 | 湘潭大学 | A kind of image classification method based on identification prediction sparse decomposition model |
| CN112597852A (en) * | 2020-12-15 | 2021-04-02 | 深圳大学 | Cell sorting method, cell sorting device, electronic apparatus, and storage medium |
| CN113869124A (en) * | 2021-08-30 | 2021-12-31 | 桂林优利特医疗电子有限公司 | Deep learning-based blood cell morphology classification method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115050024B (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147763B (en) | Video semantic segmentation method based on convolutional neural network | |
| US8600143B1 (en) | Method and system for hierarchical tissue analysis and classification | |
| KR20190043135A (en) | Systems and methods for classifying biological particles | |
| Reta et al. | Segmentation of bone marrow cell images for morphological classification of acute leukemia | |
| CN111179273A (en) | Method and system for automatically segmenting leucocyte nucleoplasm based on deep learning | |
| CN108596038B (en) | Method for identifying red blood cells in excrement by combining morphological segmentation and neural network | |
| CN108537751B (en) | Thyroid ultrasound image automatic segmentation method based on radial basis function neural network | |
| KR102122068B1 (en) | Image analyzing system and method thereof | |
| Boucher et al. | Development of a semi-automatic system for pollen recognition | |
| Mohammed et al. | Detection and classification of malaria in thin blood slide images | |
| CN111079620A (en) | Leukocyte image detection and identification model construction method based on transfer learning and application | |
| CN114945941A (en) | Non-tumor segmentation for supporting tumor detection and analysis | |
| CN111105422A (en) | Method for constructing reticulocyte classification counting model and application | |
| CN111062346A (en) | Automatic leukocyte positioning detection and classification recognition system and method | |
| Davidson et al. | Automated detection and staging of malaria parasites from cytological smears using convolutional neural networks | |
| Meimban et al. | Blood cells counting using python opencv | |
| Tantikitti et al. | Image processing for detection of dengue virus based on WBC classification and decision tree | |
| CN112784767A (en) | Cell example segmentation algorithm based on leukocyte microscopic image | |
| CN116580394A (en) | White blood cell detection method based on multi-scale fusion and deformable self-attention | |
| CN115909006A (en) | Mammary tissue image classification method and system based on convolution Transformer | |
| Yang et al. | A generalized deep learning-based framework for assistance to the human malaria diagnosis from microscopic images | |
| BENDIABDALLAH et al. | A comparison of U-net backbone architectures for the automatic white blood cells segmentation | |
| An et al. | Automated detection of tuberculosis bacilli using deep neural networks with sputum smear images | |
| Çelebi et al. | Red and white blood cell classification using Artificial Neural Networks | |
| CN115050024B (en) | Intelligent real-time identification method and system for granulocytes with interpretability |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |