CN116246700A - Tumor genotyping system and method based on hyperspectral imaging - Google Patents

Tumor genotyping system and method based on hyperspectral imaging Download PDF

Info

Publication number
CN116246700A
CN116246700A CN202211605765.9A CN202211605765A CN116246700A CN 116246700 A CN116246700 A CN 116246700A CN 202211605765 A CN202211605765 A CN 202211605765A CN 116246700 A CN116246700 A CN 116246700A
Authority
CN
China
Prior art keywords
tumor
hyperspectral
model
genotyping
spectrum
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
Application number
CN202211605765.9A
Other languages
Chinese (zh)
Inventor
李玮
赵晗竹
雷晟暄
田崇轩
安皓源
张延冰
张振磊
宋峻林
赵宇航
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong University
Original Assignee
Shandong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong University filed Critical Shandong University
Priority to CN202211605765.9A priority Critical patent/CN116246700A/en
Publication of CN116246700A publication Critical patent/CN116246700A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/58—Extraction of image or video features relating to hyperspectral data
    • 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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10036—Multispectral image; Hyperspectral 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion
    • 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
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Chemical & Material Sciences (AREA)
  • Biophysics (AREA)
  • Biotechnology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Evolutionary Biology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a tumor genotyping system and method based on hyperspectral imaging, comprising the following steps: the hyperspectral processing module is used for acquiring hyperspectral images of tumors and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral images; the model analysis module is used for obtaining a tumor genotyping result through the obtained spectrum information of each pixel point, the obtained image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is obtained by taking the spectrum information of each pixel point and the obtained image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model construction. Can realize the accurate typing of tumor genes.

Description

Tumor genotyping system and method based on hyperspectral imaging
Technical Field
The invention relates to the technical field of tumor genotyping, in particular to a tumor genotyping system and method based on hyperspectral imaging.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Most of the existing tumor genotyping methods are based on observation of tumor tissue samples and few immunohistochemical characteristics to conduct tumor genotyping, and the existing methods have the defects of few available classification characteristics and low accuracy of tumor genotyping.
The existing tumor genotyping mode needs to dye the observed pathological section of the tumor tissue, which belongs to the external forced intervention mode, and different colorants easily affect the tumor cell tissue, so that the tumor genotyping is inaccurate.
Along with the continuous innovation of single-cell sequencing technology research and high-throughput screening technology theory, various genomics data which can be obtained are increased in a geometric explosion manner, and massive data are difficult to analyze and process by an artificial naked eye screening mode or a traditional mathematical statistical method.
Therefore, the inventors believe that none of the existing methods for genotyping tumor genes can achieve accurate genotyping of tumor genes.
Disclosure of Invention
In order to solve the problems, the invention provides a tumor genotyping system and method based on hyperspectral imaging, which are used for efficiently and accurately typing tumor genes by acquiring hyperspectral images of tumors, extracting hyperspectral three-dimensional characteristic information from the hyperspectral images and identifying the three-dimensional characteristic information through a polynuclear SVM model.
In order to achieve the above purpose, the invention adopts the following technical scheme:
in a first aspect, a tumor genotyping system based on hyperspectral imaging is presented, comprising:
the hyperspectral processing module is used for acquiring hyperspectral images of tumors and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral images;
the model analysis module is used for obtaining a tumor genotyping result through the obtained spectrum information of each pixel point, the obtained image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is obtained by taking the spectrum information of each pixel point and the obtained image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model construction.
In a second aspect, a method for genotyping a tumor based on hyperspectral imaging is presented, comprising:
acquiring a hyperspectral image of a tumor, and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral image;
and acquiring a tumor genotyping result through the acquired spectrum information of each pixel point, the image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is constructed by taking the spectrum information of each pixel point and the image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model.
In a third aspect, an electronic device is provided comprising a memory and a processor and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the steps of a hyperspectral imaging-based tumor genotyping method.
In a fourth aspect, a computer readable storage medium is provided for storing computer instructions which, when executed by a processor, perform the steps of a hyperspectral imaging-based tumor genotyping method.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, the hyperspectral image of the tumor is obtained, the hyperspectral three-dimensional characteristic information containing the spectral information of each pixel point and the image information of each spectrum is extracted from the hyperspectral image, and as the three-dimensional characteristic information contains more tumor characteristic information, the accuracy of tumor genotyping is improved when the three-dimensional characteristic information is identified through the trained tumor genotyping model.
2. The tumor gene parting model is obtained by constructing the multi-core SVM model, so that the accurate identification of the three-dimensional characteristic information of the tumor hyperspectrum is realized, and the identification efficiency is improved.
Additional aspects of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application.
FIG. 1 is a block diagram of the system disclosed in example 1;
FIG. 2 is a schematic diagram showing the typing pattern of tumor genes in the system disclosed in example 1.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
Example 1
In order to achieve accurate typing of tumor genes, in this embodiment a hyperspectral imaging-based tumor genotyping system is disclosed, as shown in fig. 1, comprising: a cell pretreatment module, a hyperspectral treatment module and a model analysis module.
The cell pretreatment module comprises a puncture gun and a sample platform, wherein the puncture gun is used for acquiring tumor cells and placing the tumor cells on the sample platform.
Specifically, tumor cells are placed on a sample platform in a pathological section mode, so that subsequent hyperspectral analysis is facilitated, section staining is not needed for the case section, and genotyping analysis result deviation caused by improper operation or other factors during staining is avoided.
The hyperspectral processing module is used for acquiring hyperspectral images of tumors and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral images.
Specifically, hyperspectral processing module includes hyperspectral camera, microscopic imaging appearance and characteristic extraction module, and hyperspectral camera and microscopic imaging appearance pair use, and microscopic imaging appearance is located hyperspectral camera lower part, and hyperspectral camera passes through microscopic imaging appearance and obtains the hyperspectral image of tumour, and characteristic extraction module is connected with hyperspectral camera for draw the spectral information of every pixel point and the image information of every spectrum from hyperspectral image, the three-dimensional characteristic information of tumour hyperspectral is constituteed to the spectral information of every pixel point and the image information of every spectrum, and this three-dimensional characteristic information can comprehensive embody tumour characteristic, when utilizing this three-dimensional characteristic information to carry out tumour gene identification, has improved the rate of accuracy of tumour genotyping.
The hyperspectral camera comprises a staring type spectrometer and an imaging system, wherein the staring type spectrometer adopts a staring type spectrum scanning mode, the instrument size of the method is smaller, compared with a one-dimensional push-broom type system, the staring type filtering speed is higher, and meanwhile, complicated mechanical setting can be avoided. The microscopic imaging instrument adopts a metallographic microscopic unit, the module combines a computer with a traditional microscope, so that a visual focus and a camera focus are on the same focal plane, and the application of the microscopic imaging instrument makes the acquisition of hyperspectral digital images practical. Meanwhile, the system adopts a reflection imaging mode, so that the imaging time can be greatly reduced.
The model analysis module is used for obtaining a tumor genotyping result through the obtained spectrum information of each pixel point, the obtained image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is obtained by taking the spectrum information of each pixel point and the obtained image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model construction.
The tumor genotyping model in the model analysis module is obtained by constructing a multi-core SVM model, the multi-core SVM model adopts four Gaussian radial basis kernels, and the linear combination of the four Gaussian radial basis kernels is used as a kernel function of the multi-core SVM model, as shown in figure 2.
Specifically, the model analysis module performs contrast analysis on the obtained spectrum information of each pixel point and the obtained image information of each spectrum by using a trained tumor genotyping model, so as to realize tumor genotyping.
The multi-core SVM model is based on an original core SVM model, a traditional Linear Kernel (Linear Kernel), a polynomial Kernel (Polynomial Kernel), a Gaussian Kernel (Gaussian Kernel) and other single kernels are expanded into multiple cores, which Kernel function is not needed to be selected or parameters of which are specified according to experience or experiment are not needed, a proper Kernel is learned from an image instead, and a regularized training mode is adopted to construct a novel multi-core SVM model. The gaussian radial basis kernel has general approximation capability and is also a typical multiscale kernel. And a multi-core SVM model in the model analysis module adopts a multi-scale kernel learning mode to scale the Gaussian radial basis kernels.
The kernel function of the tumor genotyping model in the model analysis module adopts a multi-scale kernel learning method, gaussian radial basis kernels are multi-scaled, a large-scale kernel is firstly used for fitting a sample of a smooth area, then a small-scale kernel is used for fitting a sample of a relatively violent area, the result of the previous step is utilized for carrying out step-by-step optimization in the later step, and finally a better multi-kernel SVM model classification result is obtained.
Wherein, the radial basis function of Gaussian is defined as follows:
Figure BDA0003998883430000061
wherein X and X' are two samples,
Figure BDA0003998883430000062
is the squared euclidean distance between two feature vectors. Sigma is a free parameter, an equivalent but simpler definition is to set a new parameter gamma expressed as
Figure BDA0003998883430000063
The formula at this time is:
Figure BDA0003998883430000064
the process of obtaining the trained tumor genotyping model by the model analysis module is as follows:
acquiring a hyperspectral image of a tumor for training and a corresponding tumor genotyping, and extracting spectral information of each pixel point for training and image information of each spectrum from the hyperspectral image for training as training data;
and training the constructed tumor genotyping model through training data to obtain a trained tumor genotyping model.
And inputting the acquired spectrum information of each pixel point and the acquired image information of each spectrum into a trained tumor genotyping model, and outputting a tumor genotyping result.
For a given base kernel of the multi-kernel SVM model of this embodiment, four gaussian radial basis kernels are chosen here, with their linear combination as the final kernel function. By training, the weight d (weight) of each kernel in this linear combination is obtained. Due to the fusion of various cores, heterogeneous features can be taken care of; because the weight is automatically learned, which core is used and which parameter is not required to be considered, the method belongs to one of later fusion, a plurality of cores are formed by adopting different cores for different characteristics and different parameters, then the weight of each core is trained, and the optimal core function combination is selected for classification.
And constructing a new global optimization problem in an augmentation space formed by the feature spaces.
And when global optimization of the multi-scale SVM model is carried out, lagrangian coefficients are introduced, a large-scale kernel is fitted to a sample of a smooth area, then a small-scale kernel is fitted to a sample of a relatively violent area, quadratic programming is used for realizing parameter acquisition, and the later step utilizes the result of the previous step to carry out step-by-step optimization.
The multi-scale kernel sequence learning method is applied to the genotyping problem of tumors, hyperspectral image data for training are divided into 10 parts, 7 parts are used as training sets, 3 parts are used as test sets, optimization of the model is completed, the weight d of each kernel is obtained, and construction of a multi-kernel SVM model is achieved. Based on the model, the classification of the multidimensional sample of the tumor gene can be realized, the advantages of a plurality of kernel functions are exerted, the adaptability and the effectiveness of the model are greatly improved, better recognition precision and efficiency can be obtained, and the classification effect of the algorithm is better than that of the traditional support vector machine algorithm, and the algorithm has better robustness and universal applicability.
According to the embodiment, the hyperspectral image of the tumor is obtained, the hyperspectral three-dimensional characteristic information containing the spectral information of each pixel point and the image information of each spectrum is extracted from the hyperspectral image, and the three-dimensional characteristic information contains the spectral and spatial information, so that the system has more tumor characteristic information, and when the three-dimensional characteristic information is identified through the trained tumor genotyping model, the accuracy of tumor genotyping is improved.
Example 2
In this embodiment, a method of tumor genotyping based on hyperspectral imaging is disclosed, comprising:
acquiring a hyperspectral image of a tumor, and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral image;
and acquiring a tumor genotyping result through the acquired spectrum information of each pixel point, the image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is constructed by taking the spectrum information of each pixel point and the image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model.
Example 3
In this embodiment, an electronic device is disclosed comprising a memory and a processor and computer instructions stored on the memory and running on the processor that, when executed by the processor, perform the steps of a hyperspectral imaging-based tumor genotyping method disclosed in embodiment 2.
Example 4
In this embodiment, a computer readable storage medium is disclosed for storing computer instructions that, when executed by a processor, perform the steps of a hyperspectral imaging-based tumor genotyping method disclosed in embodiment 2.
Finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.

Claims (10)

1. A hyperspectral imaging-based tumor genotyping system comprising:
the hyperspectral processing module is used for acquiring hyperspectral images of tumors and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral images;
the model analysis module is used for obtaining a tumor genotyping result through the obtained spectrum information of each pixel point, the obtained image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is obtained by taking the spectrum information of each pixel point and the obtained image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model construction.
2. The hyperspectral imaging-based tumor genotyping system according to claim 1, wherein the hyperspectral processing module comprises a hyperspectral camera, a microscopic imager and a feature extraction module, wherein the hyperspectral camera obtains a hyperspectral image of a tumor through the microscopic imager, and the feature extraction module is used for extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral image.
3. The hyperspectral imaging-based tumor genotyping system of claim 1, further comprising a cell pretreatment module comprising a puncture gun and a sample platform, the puncture gun being configured to acquire tumor cells and place the tumor cells on the sample platform.
4. The hyperspectral imaging-based tumor genotyping system of claim 1, wherein the tumor genotyping model in the model analysis module is obtained by constructing a polynuclear SVM model using four gaussian radial basis kernels, and a linear combination of the four gaussian radial basis kernels as a kernel function of the polynuclear SVM model.
5. The hyperspectral imaging-based tumor genotyping system according to claim 1, wherein the kernel function of the tumor genotyping model in the model analysis module adopts a multi-scale kernel learning method to multiscale the gaussian radial basis kernels.
6. The hyperspectral imaging-based tumor genotyping system of claim 1, wherein the model analysis module obtains a trained tumor genotyping model by:
acquiring a hyperspectral image of a tumor for training and a corresponding tumor genotyping, and extracting spectral information of each pixel point for training and image information of each spectrum from the hyperspectral image for training as training data;
and training the constructed tumor genotyping model through training data to obtain a trained tumor genotyping model.
7. The hyperspectral imaging-based tumor genotyping system of claim 6, wherein the model analysis module trains the constructed tumor genotyping model in a regularized manner.
8. A method for genotyping a tumor based on hyperspectral imaging, comprising:
acquiring a hyperspectral image of a tumor, and extracting spectral information of each pixel point and image information of each spectrum from the hyperspectral image;
and acquiring a tumor genotyping result through the acquired spectrum information of each pixel point, the image information of each spectrum and the trained tumor genotyping model, wherein the tumor genotyping model is constructed by taking the spectrum information of each pixel point and the image information of each spectrum as input, taking the tumor genotyping result as output and adopting a multi-core SVM model.
9. An electronic device comprising a memory and a processor and computer instructions stored on the memory and running on the processor, which when executed by the processor, perform the steps of a hyperspectral imaging-based tumor genotyping method as claimed in claim 8.
10. A computer readable storage medium storing computer instructions which, when executed by a processor, perform the steps of a hyperspectral imaging-based tumor genotyping method as claimed in claim 8.
CN202211605765.9A 2022-12-14 2022-12-14 Tumor genotyping system and method based on hyperspectral imaging Pending CN116246700A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211605765.9A CN116246700A (en) 2022-12-14 2022-12-14 Tumor genotyping system and method based on hyperspectral imaging

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211605765.9A CN116246700A (en) 2022-12-14 2022-12-14 Tumor genotyping system and method based on hyperspectral imaging

Publications (1)

Publication Number Publication Date
CN116246700A true CN116246700A (en) 2023-06-09

Family

ID=86623175

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211605765.9A Pending CN116246700A (en) 2022-12-14 2022-12-14 Tumor genotyping system and method based on hyperspectral imaging

Country Status (1)

Country Link
CN (1) CN116246700A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117953970A (en) * 2024-03-27 2024-04-30 山东大学 Lung cancer polygene detection method and system based on hyperspectral image

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117953970A (en) * 2024-03-27 2024-04-30 山东大学 Lung cancer polygene detection method and system based on hyperspectral image
CN117953970B (en) * 2024-03-27 2024-06-11 山东大学 Lung cancer polygene detection method and system based on hyperspectral image

Similar Documents

Publication Publication Date Title
CN108364288B (en) Segmentation method and device for breast cancer pathological image
CN110472676A (en) Stomach morning cancerous tissue image classification system based on deep neural network
CN109993236B (en) One-shot Simese convolutional neural network-based small-sample Manchu matching method
CN110633758A (en) Method for detecting and locating cancer region aiming at small sample or sample unbalance
CN111161272B (en) Embryo tissue segmentation method based on generation of confrontation network
CN111028923B (en) Digital pathological image staining normalization method, electronic device and storage medium
CN111080592B (en) Rib extraction method and device based on deep learning
Wang et al. A generalizable and robust deep learning algorithm for mitosis detection in multicenter breast histopathological images
Kheradmand et al. Inner cell mass segmentation in human hmc embryo images using fully convolutional network
CN111047559B (en) Method for rapidly detecting abnormal area of digital pathological section
CN113393443A (en) HE pathological image cell nucleus segmentation method and system
CN115496720A (en) Gastrointestinal cancer pathological image segmentation method based on ViT mechanism model and related equipment
CN116246700A (en) Tumor genotyping system and method based on hyperspectral imaging
Zhang et al. Constrained multi-scale dense connections for accurate biomedical image segmentation
Juhong et al. Super-resolution and segmentation deep learning for breast cancer histopathology image analysis
CN114864075A (en) Glioma grade analysis method and device based on pathological image
CN114387264A (en) HE staining pathological image data expansion and enhancement method
Huang et al. Evaluations of deep learning methods for pathology image classification
Ning et al. Multiscale context-cascaded ensemble framework (MsC 2 EF): application to breast histopathological image
CN112508860B (en) Artificial intelligence interpretation method and system for positive check of immunohistochemical image
WO2021067507A1 (en) Building computational transfer functions on 3d light microscopy images using deep learning
Sayın et al. Comparative Analysis of Deep Learning Architectures for Breast Cancer Diagnosis Using the BreaKHis Dataset
CN113177602B (en) Image classification method, device, electronic equipment and storage medium
CN113222887A (en) Deep learning-based nano-iron labeled neural stem cell tracing method
Sapkota et al. Automatic muscle perimysium annotation using deep convolutional neural network

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