Angiopoietin detection method and device, electronic equipment and storage medium

Abstract

The invention discloses an angiopoietin detection method and device, electronic equipment and a storage medium, and relates to the technical field of image processing, and the method comprises the steps: obtaining hematoxylin-eosin staining pathological images of different hepatocellular carcinoma patient tissues; segmenting cancer tissue areas in each pathological image, and randomly selecting the cancer tissue areas as sub-images; extracting a plurality of features of each sub-image, and screening the features to obtain an optimal feature subset; constructing a pathological omics model based on the optimal feature subset; and calculating the angiopoietin expression probability of the tissue of the corresponding hepatocellular carcinoma patient according to the characteristics by using a pathomics model, and detecting the expression level of the angiopoietin according to the angiopoietin expression probability. According to the method, the rich features of the pathological image are obtained, then the pathological omics model is constructed, the corresponding angiopoietin expression probability is predicted for each feature, the expression level of the angiopoietin can be objectively and truly detected based on the pathological image, and the detection accuracy of the angiopoietin is improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an angiogenin detection method, device, electronic device and storage medium. Background Art

[0002] Currently, the detection of angiopoietin 2 (ANGPT2) expression levels includes the following schemes: real-time detection of peripheral blood cytokines, but it cannot reflect the situation of tumor substance; detection based on fresh tissue specimens, but there are difficulties in collecting tissue specimens and the detection accuracy is affected by the antibodies of the detection operator and the subject; detection based on paraffin tissue specimens, the detection accuracy is also affected by the antibodies of the detection operator and the subject. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to provide an angiogenin detection method, device, electronic device and storage medium to improve the detection accuracy of angiogenin.

[0004] To achieve the above object, one aspect of the embodiments of the present application provides a method for detecting angiogenin, the method comprising the following steps:

[0005] Obtain hematoxylin-eosin stained pathological images of tissues from different hepatocellular carcinoma patients;

[0006] Segmenting the cancerous tissue area in each of the pathological images and randomly selecting them as sub-images;

[0007] Extracting multiple features of each of the sub-images, and selecting an optimal feature subset from each of the features;

[0008] constructing a pathomic model based on the optimal feature subset;

[0009] The pathological omics model is used to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the characteristics, and the expression level of angiogenin is detected according to the angiogenin expression probability.

[0010] In some embodiments, the step of obtaining hematoxylin-eosin stained pathological images of different hepatocellular carcinoma patient tissues comprises the following steps:

[0011] Obtain pathological images of hematoxylin-eosin staining of formalin-fixed and paraffin-embedded tissues from different hepatocellular carcinoma patients.

[0012] In some embodiments, segmenting the cancerous tissue regions in each of the pathological images and randomly selecting them as sub-images comprises the following steps:

[0013] Segmenting a plurality of cancerous tissue regions from each of the pathological images according to a preset threshold value using an image binarization segmentation threshold algorithm;

[0014] Segmenting each of the cancerous tissue regions into a plurality of sub-images of preset sizes;

[0015] A plurality of representative sub-images are randomly selected for each pathological image.

[0016] In some embodiments, extracting a plurality of features of each of the sub-images comprises the following steps:

[0017] Extract the original features and high-order features of each sub-image as the features; wherein the original features include first-order statistics, grayscale co-occurrence matrix, grayscale run length matrix, grayscale size area, adjacent grayscale difference matrix, and grayscale dependency matrix; the high-order features are obtained by wavelet transforming the original features.

[0018] In some embodiments, the screening of the optimal feature subset from each of the features comprises the following steps:

[0019] Calculate the information gain mean between each of the features as the feature category correlation coefficient;

[0020] Calculate the sum of mutual information between each of the features, and then divide the sum of mutual information by the square of the total number of features to obtain a feature redundancy coefficient;

[0021] Screening a preset number of features from each of the features according to the feature category correlation coefficient and the feature redundancy coefficient as a candidate feature subset;

[0022] Calculate the importance score of each feature in the candidate feature subset;

[0023] The features ranked low in importance scores are removed, and the remaining features in the candidate feature subset are used as the optimal feature subset.

[0024] In some embodiments, constructing a pathology omics model based on the optimal feature subset comprises the following steps:

[0025] Dividing a training set from images of hepatocellular carcinoma patients obtained from a professional database, and extracting the optimal feature subset from the training set;

[0026] Constructing the pathogenomics model based on the optimal feature subset extracted from the training set;

[0027] Calculating the angiogenin expression probability corresponding to each of the features in the training set using the pathological omics model as a predicted value;

[0028] Calculating the residual and gradient between the predicted value and the actual value;

[0029] Constructing a weak learner according to the residual, and fitting the weak learner with the residual to train the newly added weak learner along the negative direction of the gradient;

[0030] Adding the weak learner to the pathological omics model as a new pathological omics model;

[0031] The angiogenin expression probability corresponding to each of the features in the training set calculated by using the pathological omics model is returned as a predicted value until a preset condition is met.

[0032] In some embodiments, the method further comprises the following steps:

[0033] The R language survminer package was used to perform binary classification on the angiogenin expression probabilities output by the pathological omics model.

[0034] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application provides a device for detecting angiogenin, the device comprising:

[0035] An image acquisition unit, used for acquiring hematoxylin-eosin staining pathological images of tissues of different hepatocellular carcinoma patients;

[0036] An image segmentation unit, used for segmenting cancerous tissue regions in each of the pathological images and randomly selecting them as sub-images;

[0037] A feature processing unit, used for extracting multiple features of each of the sub-images, and screening out an optimal feature subset from each of the features;

[0038] A model building unit, used for building a pathology omics model based on the optimal feature subset;

[0039] An expression level detection unit is used to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the characteristics using the pathological omics model, and detect the expression level of angiogenin according to the angiogenin expression probability.

[0040] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.

[0041] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0042] The embodiments of the present application include at least the following beneficial effects:

[0043] This application can obtain hematoxylin-eosin stained pathological images of different hepatocellular carcinoma patient tissues; segment the cancerous tissue areas in each pathological image and randomly select them as sub-images; extract multiple features of each sub-image, and screen the optimal feature subset from each feature; construct a pathological omics model based on the optimal feature subset; use the pathological omics model to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the features, and detect the expression level of angiogenin according to the angiogenin expression probability. By obtaining the rich features of the pathological image and then constructing the pathological omics model to predict the corresponding angiogenin expression probability for each feature, the expression level of angiogenin can be objectively and truly detected based on the pathological image. Compared with the existing technology, it can reduce the error caused by the operator and the subject's antibodies, and improve the detection accuracy of angiogenin. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A schematic diagram of a flow chart of a method for detecting angiogenin provided in an embodiment of the present application;

[0046] Figure 2 An exemplary flow chart of a method for detecting angiogenin provided in an embodiment of the present application;

[0047] Figure 3 An example diagram of the optimal feature subset provided in the embodiment of the present application;

[0048] Figure 4 The ROC curves, calibration curves and decision curve diagrams of the training set, test set and external test set provided in the embodiments of the present application;

[0049] Figure 5 A schematic diagram of the structure of an angiogenin detection device provided in an embodiment of the present application;

[0050] Figure 6 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0052] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0053] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0055] The embodiment of the present application provides a method for detecting angiogenin, which relates to the field of image processing technology. The angiogenin detection method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services. The cloud server, the server can also be a node server in a blockchain network; the software can be an application that implements the angiogenin detection method, etc., but is not limited to the above forms.

[0056] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0057] Reference Figure 1 The present application embodiment provides a method for detecting angiogenin, which may include but is not limited to S100 to S140, as follows:

[0058] S100: Obtain hematoxylin-eosin stained pathological images of tissues from different hepatocellular carcinoma patients.

[0059] Specifically, this embodiment can obtain multiple HE (Hematoxylin-Eosin) stained sections, ie, pathological images, for different tissues of patients with hepatocellular carcinoma.

[0060] In order to enrich the pathological images and obtain different pathological images, the present embodiment can obtain multiple pathological images from a professional database and obtain pathological images obtained by examining the patient in real time.

[0061] As a further implementation, S100 may include:

[0062] Obtain pathological images of hematoxylin-eosin staining of formalin-fixed and paraffin-embedded tissues from different hepatocellular carcinoma patients.

[0063] Specifically, the tissues of hepatocellular carcinoma patients were fixed in formalin and embedded in paraffin, and then the hematoxylin-eosin stained pathological images of the tissues of hepatocellular carcinoma patients were obtained.

[0064] S110: Segment the cancerous tissue area in each of the pathological images and randomly select them as sub-images.

[0065] It can be understood that the area containing the tissue slice in the pathological image is used as the cancerous tissue area, and the remaining area is used as the background area. The cancerous tissue area is segmented from the pathological image as a sub-image.

[0066] Furthermore, S110 may include S111 to S113:

[0067] S111: Segmenting a plurality of cancerous tissue regions from each of the pathological images according to a preset threshold value using an image binarization segmentation threshold algorithm.

[0068] Exemplarily, this embodiment may use the OTSU algorithm as an image binarization segmentation threshold algorithm to segment the pathological image.

[0069] S112: Segment each of the cancerous tissue regions into a plurality of sub-images of preset sizes.

[0070] To unify the image processing standard, this embodiment can segment the cancer tissue area into multiple sub-images of preset sizes.

[0071] S113: Randomly selecting a plurality of representative sub-images for each pathological image.

[0072] Exemplarily, each sub-image is manually reviewed to exclude sub-images with poor image quality (contamination, blurred images, blank areas exceeding 50%). More specifically, 10 sub-images are randomly selected from each pathological image for subsequent analysis.

[0073] S120: extracting multiple features of each of the sub-images, and screening out an optimal feature subset from each of the features.

[0074] Specifically, the pathological image of the tissue contains rich tumor genetic information and can provide a large number of pathological features, such as tumor tissue structure, cell morphology, tumor microenvironment, immune cell infiltration, tumor differentiation degree, microvascular invasion degree, etc. This embodiment can extract the above features from the pathological image, screen the extracted features, and then use the screened features as the optimal feature subset.

[0075] Furthermore, extracting a plurality of features of each of the sub-images comprises the following steps:

[0076] Extract the original features and high-order features of each sub-image as the features; wherein the original features include first-order statistics, grayscale co-occurrence matrix, grayscale run length matrix, grayscale size area, adjacent grayscale difference matrix, and grayscale dependency matrix; the high-order features are obtained by wavelet transforming the original features.

[0077] Specifically, first-order statistics (First Order Statistics): describe the distribution of voxel intensities within the image region defined by the mask through common and basic indicators (18, the standard deviation is consistent with the variance, and the variance is taken).

[0078] Gray Level Co-occurrence Matrix (24): describes the second-order joint probability function of the image region constrained by the mask.

[0079] Gray Level Run Length Matrix (16): quantized gray level runs, which are defined as the number of pixels of consecutive pixels with the same gray level value.

[0080] Gray Level Size Zone Matrix (16): quantizes the grayscale area in the image.

[0081] Neigbouring Gray Tone Difference Matrix (5): quantifies the difference between the gray value within the distance δ and the average gray value of its neighboring area.

[0082] Gray Level Dependence Matrix (14): quantifies the gray level dependency in the image. Gray level dependency is defined as the number of connected voxels that depend on the central voxel within a distance δ.

[0083] In addition, this embodiment can also extract high-order features of pathological images. The high-order features are obtained by transforming the original features through wavelet. LL, LH, HL, and HH are four transformation methods, namely Wavelet (LL, LH, HL, HH). Finally, 93 (Original features) + 93*4 (high-order features) = 465 features are obtained.

[0084] Furthermore, the optimal feature subset is screened from each of the features, including S121 to S125:

[0085] S121: Calculate the information gain mean between each of the features as the feature category correlation coefficient;

[0086] S122: Calculate the sum of mutual information between the features, and then divide the sum of mutual information by the square of the total number of features to obtain a feature redundancy coefficient;

[0087] S123: Screening a preset number of features from each of the features according to the feature category correlation coefficient and the feature redundancy coefficient as a candidate feature subset;

[0088] S124: Calculate the importance score of each feature in the candidate feature subset;

[0089] S125: removing the plurality of features ranked low in importance scores, and using the remaining features in the candidate feature subset as the optimal feature subset.

[0090] It can be understood that in this embodiment, the correlation and redundancy between the various features are first calculated, and then a candidate feature subset is obtained by preliminary screening, and then the importance score of each feature in the candidate feature subset is calculated. Features with higher importance scores contribute more to the prediction of angiogenin and are therefore retained, while features with lower importance scores can be eliminated.

[0091] S130: constructing a pathology omics model based on the optimal feature subset.

[0092] Specifically, the optimal feature subset may include features that are highly correlated with angiogenin, and a pathological omics model capable of predicting the probability of angiogenin expression is constructed based on the optimal feature subset.

[0093] Further, S130 may include S131 to S137:

[0094] S131: dividing a training set from images of hepatocellular carcinoma patients obtained from a professional database, and extracting the optimal feature subset from the training set;

[0095] S132: constructing the pathogenomics model based on the optimal feature subset extracted from the training set;

[0096] S133: using the pathological omics model to calculate the angiogenin expression probability corresponding to each of the features in the training set as a predicted value;

[0097] S134: Calculate the residual and gradient between the predicted value and the actual value;

[0098] S135: constructing a weak learner according to the residual, and fitting the weak learner with the residual to train the newly added weak learner along the negative direction of the gradient;

[0099] S136: adding the weak learner to the pathological omics model as a new pathological omics model;

[0100] S137: Returning to the step of calculating the angiogenin expression probability corresponding to each of the features in the training set using the pathological omics model as a predicted value until a preset condition is met.

[0101] Specifically, the preset conditions for stopping training the pathology omics model may include reaching a preset number of times or the performance parameters of the pathology omics model are no longer significantly improved, that is, the performance parameters are improved by a small or stable range.

[0102] S140: Calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue using the pathological omics model according to the characteristics, and detect the expression level of angiogenin according to the angiogenin expression probability.

[0103] Specifically, the higher the angiogenin expression probability is, the higher the probability of angiogenin existing in the tissue of the hepatocellular carcinoma patient is, and thus the expression level of angiogenin can be detected based on the angiogenin expression probability.

[0104] After detecting angiogenin, the embodiment of the present application may further include a step of classifying the detection results, specifically including:

[0105] S150: using the survminer package of R language to perform binary classification on the angiogenin expression probabilities output by the pathological omics model.

[0106] Specifically, the pathological omics model obtained according to this embodiment outputs the expression probability of each angiogenin as a prediction value, and each prediction value is classified into two categories based on the minimum P value calculated using the survminer package of the R language.

[0107] Next, the solution of the embodiment of the present application will be introduced and explained in detail with reference to specific application examples.

[0108] Figure 2 For an example flow chart of an angiogenin detection method provided in this embodiment, refer to Figure 2 For illustration, this embodiment may include Step 1 to Step 5:

[0109] Step 1: Obtain hematoxylin-eosin (HE) stained sections of tumor tissue from patients with hepatocellular carcinoma to provide a basis for subsequent digital processing and analysis.

[0110] Specifically, 267 pathological images of hepatocellular carcinoma patients were downloaded from professional databases, and 91 tissue section images of hepatocellular carcinoma patients were obtained from actual patient cases: formalin-fixed and paraffin-embedded pathological tissue sections, SVS format, and the maximum magnification was 20× or 40× HE-stained tissue pathological images (20× or 40× magnification).

[0111] The images downloaded from the professional database were randomly divided into training set and test set in a ratio of 7:3. Among them, 70% of the data in the training set were used to train the pathology model; the remaining 30% of the data were used as the test set. The baseline conditions of the patients in the training set and the test set were similar, and the groups were comparable. Pathological images obtained from actual patient cases were used as the external test set.

[0112] Step 2: Process and segment the acquired pathological images to obtain subimages of each pathological image for subsequent feature extraction.

[0113] Specifically, the OTSU algorithm is used to obtain the cancerous tissue region of the pathological image, that is, the tissue region. The OTSU algorithm, also known as the maximum inter-class variance method, is an image binarization segmentation threshold algorithm that uses a threshold to segment an image into a background region and a tissue region required for research.

[0114] Segment pathological tissue images into sub-images: Segment the 40× pathological image into multiple 1024×1024 pixel sub-images; segment the 20× pathological image into multiple 512×512 pixel sub-images and upsample them to 1024×1024 pixels.

[0115] Screening sub-images: Optionally, manual review is performed to exclude sub-images with poor image quality (contamination, blurred images, blank areas exceeding 50%). 10 sub-images are randomly selected from each pathology image for subsequent analysis.

[0116] Step 3: Normalize the sub-image and extract features, and use the algorithm to filter out the optimal feature subset.

[0117] Specifically, this embodiment can perform pathological image feature extraction. The PyRadiomics open source package is used to standardize the sub-images and extract 93 original features (including first-order features and second-order features), and high-order features (Wavelet (LL, LH, HL, HH)) are extracted, and a total of 465 features are obtained. After extracting features from the 10 sub-images of each patient's pathological image, the corresponding average value is taken as the pathological genomics feature of each sample for subsequent data analysis.

[0118] Among them, the Original feature types include:

[0119] First Order Statistics: Describes the distribution of voxel intensities within the image region defined by the mask using common and basic indicators (18, standard deviation is consistent with variance, variance is taken).

[0120] Gray Level Co-occurrence Matrix (24): describes the second-order joint probability function of the image region constrained by the mask.

[0121] Gray Level Run Length Matrix (16): quantized gray level runs, which are defined as the number of pixels of consecutive pixels with the same gray level value.

[0122] Gray Level Size Zone Matrix (16): quantizes the grayscale area in the image.

[0123] Neigbouring Gray Tone Difference Matrix (5): quantifies the difference between the gray value within the distance δ and the average gray value of its neighboring area.

[0124] Gray Level Dependence Matrix (14): quantifies the gray level dependency in the image. Gray level dependency is defined as the number of connected voxels that depend on the central voxel within a distance δ.

[0125] The high-order features are obtained by transforming the original features through wavelet transformation. LL, LH, HL, and HH are four transformation methods, namely Wavelet (LL, LH, HL, HH). Finally, 93 (original features) + 93*4 (high-order features) = 465 features are obtained.

[0126] The steps of screening the optimal feature subset include: using an algorithm to screen the optimal feature subset to help the pathology omics model reduce overfitting of the training data. First, the mRMR (maximum relevance, minimum redundancy) algorithm is used to select the top 30 features of importance. The maximum relevance and minimum redundancy algorithm mRMR (Maximum relevance relevance, minimum redundancy redundancy) screens features, which not only takes into account the correlation between the features and the variables to be predicted, but also takes into account the correlation between features. The metric used is mutual information. For the mRMR method, the correlation between the feature subset and the category is calculated by the mean information gain of each feature and the category, while the redundancy between features is calculated by adding the mutual information between features and dividing it by the square of the total number of features in the optimal feature subset.

[0127] Subsequently, the 30 features obtained are filtered using the RFE (Recursive feature elimination) algorithm. RFE (Recursive feature elimination elimination) feature screening sorts the features before modeling, and unimportant features are eliminated one by one. The goal is to find the optimal feature subset that can be used to generate an accurate model. The pathology omics model is trained multiple times, and n features with low importance scores are deleted after each training. Then the new features are trained again, and the importance scores of the features are obtained again. Then, n features with low importance scores are deleted again until the optimal feature subset is obtained. Exemplarily, Figure 3 An example diagram of the optimal feature subset provided for this embodiment.

[0128] Step 4: Based on the optimal feature subset, a pathology omics model is constructed, and machine learning algorithms are used to conduct in-depth analysis of pathological features to improve the generalization ability and prediction accuracy of the pathology omics model.

[0129] Specifically, the features screened by the mRMR-RFE algorithm were used to construct a pathological omics model in the training set through the Gradient Boosting Machine algorithm.

[0130] The specific process of the gradient enhancement model algorithm is as follows:

[0131] Step 1: Initialization - Set the initial prediction value to a constant (e.g., all prediction values ​​in a regression problem are the target mean of the training data).

[0132] Step 2: Iteration - Calculate the residual or gradient between the predicted result of the current pathogenomics model and the actual target value.

[0133] A new weak learner (such as a decision tree) is constructed based on the residuals to fit the residuals, that is, the parameters of the pathomic model are updated along the negative gradient direction.

[0134] Step 3: Update the pathomic model - add the new weak learner to the current pathomic model to obtain a new pathomic model.

[0135] Step 4: Adjust the learning rate (shrinkage) to control the learning speed of each step and reduce overfitting.

[0136] The above steps are repeated until a predetermined number of iterations is reached or the performance parameters of the pathomic model are no longer significantly improved.

[0137] Step 5: The pathomics model outputs the probability pathomics score (PS) for predicting the expression level of angiopoietin 2 (ANGPT2) (mRNA level or protein level detected by immunohistochemistry), and the pathological images are grouped to predict the ANGPT2 expression level.

[0138] Output of prediction value: The finally trained pathomics model is used to generate a probability of high molecular weight expression (0-100%) according to the characteristic values ​​of different pathological images, namely the pathomics score (PS). PS is the probability of the pathomics model predicting the expression level of ANGPT2, and its value ranges from 0 to 1. The closer the PS value is to 0, the lower the ANGPT2 expression level predicted by the pathomics model; conversely, the closer the PS value is to 1, the higher the ANGPT2 expression level predicted by the pathomics model.

[0139] Division of prediction values: The pathology score (Pathomics score) calculated by the pathology model was merged with the clinical data, and the cutoff value of the pathology score, i.e., the classification threshold, was calculated through the R package (mainly ggplot2 [version 3.3.6]) "survminer". Due to different levels of prediction, the pathology images downloaded from the professional database are the mRNA levels of ANGPT2, and the cutoff value of the predicted value of the pathology score based on the minimum P value GBM (Gradient Boosting Machines) model is 0.512; the external test set is based on the immunohistochemical protein level of ANGPT2, and the cutoff value of the predicted value of the pathology score using the median GBM model is 0.500. According to the cutoff value, the pathology images are divided into two categories of high / low expression levels of angiopoietin.

[0140] Demonstration of the predictive performance of the pathomic model:

[0141] A pathology omics model was established and verified based on pathology images of professional databases and actual cases. The pathology omics model showed high predictive accuracy in the training set, test set and external test set.

[0142] Specific evaluation indicators of the performance of pathology omics models include: (1) ROC curve: showing the true positive rate (sensitivity) and false positive rate (1-specificity) of the pathology omics model at the set threshold. (2) Area under the ROC curve: measures the ability of the pathology omics model to distinguish different efficacy prediction groups. The larger the ROC-AUC, the larger the area under the curve, the more the curve bulges toward the upper left corner, and the better the effect of the pathology omics model. The AUC value of a perfect classifier is 1, while the AUC value of a model with no distinguishing ability is close to 0.5. (3) Draw a calibration curve to evaluate the calibration of the pathology omics model; use the Brier score to quantify the comprehensive performance of the imaging omics prediction model. The smaller the value, the better the consistency of the pathology omics model prediction; (4) Decision curve (DCA) to show the clinical benefit of the pathology omics prediction model. Figure 4 ROC curves, calibration curves, and decision curves for the training set, test set, and external test set. Figure 4 AC: ROC shows the prognostic performance of the model in the training, test, and external test sets. The AUC value of the model is 0.811, the AUC of the test set is 0.726, and the AUC of the external test set is 0.710. Figure 4 DF in : The calibration curve shows that the predicted probabilities in the training, test, and external test sets are consistent with the model for the actual states. Figure 4 GI in: DCA showed clinical benefit of the prediction model in training, test, and external test sets.

[0143] The above results not only prove that the pathological omics model has a good predictive effect, but also has a good consistency between the predicted probability and the true value of whether the gene is highly expressed, and has high clinical practicality and extensiveness.

[0144] Beneficial effects of this embodiment:

[0145] (1) Accurate prediction: It can fully extract tumor feature information contained in pathological images. The analysis results are not affected by individual differences such as age and gender. It overcomes the limitation of existing methods that fail to fully consider tumor heterogeneity, and at the same time improves the objectivity and accuracy of prediction. (2) Efficient image segmentation: In the process of pathological omics analysis, the OTSU algorithm is used to accurately segment tumor tissue in pathological images, and automatic delineation of tumor areas is achieved, laying a solid foundation for in-depth research on the morphological characteristics of tumors and predicting patient survival. (3) Good generalization: Using the features selected by the mRMR-RFE algorithm, the GBM algorithm is used to construct a pathological omics model in the training set, demonstrating the versatility and powerful feature extraction capabilities of the pathological omics model. Through external verification, the trained model can adapt to specific pathological image analysis tasks and show good performance on new data sets.

[0146] This embodiment is able to comprehensively extract the morphological features of the tumor, taking into account tumor heterogeneity, while existing technologies may not fully consider this complexity of the tumor. In response to the computational challenges brought by the high number of pixels in pathological tissue images, this embodiment uses algorithms to efficiently distinguish tumors from background areas, which not only simplifies the analysis process, but also significantly reduces the computational burden of high-resolution histopathology analysis. The pathomic model was trained on pathological images from professional databases and passed internal and external validation, demonstrating its potential for wide application and high predictive accuracy in a variety of clinical settings.

[0147] In addition to the pathological image processing scheme of the embodiment of the present application, the following alternatives can also be considered to achieve the detection of angiogenin: (1) Genomic analysis: It involves gene sequencing of tumor tissues of liver cancer patients and analyzing gene expression patterns, which directly reflects the molecular expression level. This scheme can directly reveal the molecular characteristics of the tumor. (2) Imaging genomics technology: Using imaging technologies such as computed tomography or magnetic resonance imaging, combined with imaging genomics analysis, features are extracted from image data to predict molecular expression. This scheme can provide morphological information of the tumor. (3) Multimodal data fusion: Combining pathological genomics data with imaging genomics and genomics data, and improving the accuracy of the prediction model through integrated learning or data fusion technology. This scheme may improve the accuracy of the prediction.

[0148] Although the above alternatives have their own advantages, this embodiment provides a more direct and cost-effective solution through pathological omics analysis, which is particularly suitable for detection scenarios lacking high-end imaging equipment or gene sequencing technology. In addition, this embodiment has shown higher prediction accuracy than existing technologies in internal and external tests and is easy to implement.

[0149] Reference Figure 5 The present application also provides an angiogenin detection device, which can implement the above-mentioned angiogenin detection method. The device includes:

[0150] An image acquisition unit, used for acquiring hematoxylin-eosin staining pathological images of tissues of different hepatocellular carcinoma patients;

[0151] An image segmentation unit, used for segmenting cancerous tissue regions in each of the pathological images and randomly selecting them as sub-images;

[0152] A feature processing unit, used for extracting multiple features of each of the sub-images, and screening out an optimal feature subset from each of the features;

[0153] A model building unit, used for building a pathology omics model based on the optimal feature subset;

[0154] An expression level detection unit is used to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the characteristics using the pathological omics model, and detect the expression level of angiogenin according to the angiogenin expression probability.

[0155] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0156] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned angiogenin detection method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0157] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0158] See also Figure 6 , Figure 6 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0159] The processor 601 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0160] The memory 602 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 602 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 602, and the processor 601 is called to execute the angiogenin detection method of the embodiment of this application;

[0161] Input / output interface 603, used to implement information input and output;

[0162] Communication interface 604, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0163] Bus 605 , which transmits information between various components of the device (e.g., processor 601 , memory 602 , input / output interface 603 , and communication interface 604 );

[0164] The processor 601 , the memory 602 , the input / output interface 603 and the communication interface 604 are connected to each other in communication within the device via a bus 605 .

[0165] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned angiogenin detection method is implemented.

[0166] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0167] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0168] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0169] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0170] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0171] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0172] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0173] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0174] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0175] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0177] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0178] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for detecting angiogenin, characterized in that: The method comprises the following steps: Obtain hematoxylin-eosin stained pathological images of tissues from different hepatocellular carcinoma patients; Segmenting the cancerous tissue area in each of the pathological images and randomly selecting them as sub-images; Extracting multiple features of each of the sub-images, and selecting an optimal feature subset from each of the features; constructing a pathomic model based on the optimal feature subset; The pathological omics model is used to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the characteristics, and the expression level of angiogenin is detected according to the angiogenin expression probability.

2. The method for detecting angiogenin according to claim 1, characterized in that: The method of obtaining hematoxylin-eosin stained pathological images of tissues of different hepatocellular carcinoma patients comprises the following steps: Obtain pathological images of hematoxylin-eosin staining of formalin-fixed and paraffin-embedded tissues from different hepatocellular carcinoma patients.

3. The method for detecting angiogenin according to claim 1, characterized in that: The segmenting of the cancerous tissue regions in each of the pathological images and randomly selecting them as sub-images comprises the following steps: Segmenting a plurality of cancerous tissue regions from each of the pathological images according to a preset threshold value using an image binarization segmentation threshold algorithm; Segmenting each of the cancerous tissue regions into a plurality of sub-images of preset sizes; A plurality of representative sub-images are randomly selected for each pathological image.

4. The method for detecting angiogenin according to claim 1, characterized in that: The step of extracting a plurality of features from each of the sub-images comprises the following steps: Extract the original features and high-order features of each sub-image as the features; wherein the original features include first-order statistics, grayscale co-occurrence matrix, grayscale run length matrix, grayscale size area, adjacent grayscale difference matrix, and grayscale dependency matrix; the high-order features are obtained by wavelet transforming the original features.

5. The method for detecting angiogenin according to claim 1, characterized in that: The step of selecting the optimal feature subset from each of the features comprises the following steps: Calculate the information gain mean between each of the features as the feature category correlation coefficient; Calculate the sum of mutual information between each of the features, and then divide the sum of mutual information by the square of the total number of features to obtain a feature redundancy coefficient; Screening a preset number of features from each of the features according to the feature category correlation coefficient and the feature redundancy coefficient as a candidate feature subset; Calculate the importance score of each feature in the candidate feature subset; The features ranked low in importance scores are removed, and the remaining features in the candidate feature subset are used as the optimal feature subset.

6. The method for detecting angiogenin according to claim 1, characterized in that: The step of constructing a pathology omics model based on the optimal feature subset comprises the following steps: Dividing a training set from images of hepatocellular carcinoma patients obtained from a professional database, and extracting the optimal feature subset from the training set; Constructing the pathogenomics model based on the optimal feature subset extracted from the training set; Calculating the angiogenin expression probability corresponding to each of the features in the training set using the pathological omics model as a predicted value; Calculating the residual and gradient between the predicted value and the actual value; Constructing a weak learner according to the residual, and fitting the weak learner with the residual to train the newly added weak learner along the negative direction of the gradient; Adding the weak learner to the pathological omics model as a new pathological omics model; The angiogenin expression probability corresponding to each of the features in the training set calculated by using the pathological omics model is returned as a predicted value until a preset condition is met.

7. A method for detecting angiogenin according to any one of claims 1 to 6, characterized in that: The method further comprises the following steps: The R language survminer package was used to perform binary classification on the angiogenin expression probabilities output by the pathological omics model.

8. An angiogenin detection device, characterized in that: The device comprises: An image acquisition unit, used for acquiring hematoxylin-eosin staining pathological images of tissues of different hepatocellular carcinoma patients; An image segmentation unit, used for segmenting cancerous tissue regions in each of the pathological images and randomly selecting them as sub-images; A feature processing unit, used for extracting multiple features of each of the sub-images, and screening out an optimal feature subset from each of the features; A model building unit, used for building a pathology omics model based on the optimal feature subset; An expression level detection unit is used to calculate the angiogenin expression probability of the corresponding hepatocellular carcinoma patient tissue according to the characteristics using the pathological omics model, and detect the expression level of angiogenin according to the angiogenin expression probability.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images