Lung cancer patient prognosis evaluation method and system in combination with pathological image and blood index

Abstract

The invention discloses a lung cancer patient prognosis evaluation method and system combining pathological images and blood indexes. The method comprises the steps that reference pathological features and reference blood clinical indexes of a lung cancer patient in different clinical stages are obtained, and a lung cancer stage evaluation model is constructed; acquiring a real-time pathological image and a blood detection report of the target patient, and determining a current lung cancer development stage of the target patient through the lung cancer stage evaluation model; determining the risk attribute of the target patient based on the current lung cancer development stage and determining the blood lymphocyte proportion of the target patient; determining a key blood index of the target patient according to the blood lymphocyte proportion, and outputting a prognosis risk score and a lifetime prediction result of the target patient through a preset survival analysis model according to the key blood index and the multi-scale morphological characteristics. The objectivity, the high precision and the stability of the prediction result are ensured, the interference of human subjective factors is eliminated, and the reliability is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lung cancer prognosis evaluation, and in particular to a lung cancer patient prognosis evaluation method and system combining pathological images and blood indicators. BACKGROUND

[0002] At present, lung cancer is one of the malignant tumors with the fastest growth of incidence and mortality, and the greatest threat to human health and life. In the past 50 years, many countries have reported that the incidence and mortality of lung cancer have increased significantly. The incidence and mortality of male lung cancer account for the first place of all malignant tumors, and the incidence of female lung cancer accounts for the second place, and the mortality accounts for the second place. Therefore, the prognosis evaluation of lung cancer is crucial for treatment decision and survival prediction. At present, the clinical mainly relies on the biopsy sample of the physician based on the unit time point for static evaluation, which has the following problems: the evaluation process mainly depends on artificial experience, it is difficult to capture the dynamic evolution and treatment response of the tumor, the time sequence change information of the patient state in the treatment process is not integrated, so as to ensure the accuracy and reliability of the prognosis survival prediction, and reduce the practicability and stability. SUMMARY

[0003] In view of the above problems, the present application provides a lung cancer patient prognosis evaluation method and system combining pathological images and blood indicators to solve the problems of the evaluation process mainly depending on artificial experience, being difficult to capture the dynamic evolution and treatment response of the tumor, not integrating the time sequence change information of the patient state in the treatment process, so as to ensure the accuracy and reliability of the prognosis survival prediction, and reducing the practicability and stability.

[0004] A lung cancer patient prognosis evaluation method combining pathological images and blood indicators, comprising the following steps:

[0005] Obtaining reference pathological features and reference blood clinical indicators of lung cancer patients at different clinical stages, and constructing a lung cancer stage evaluation model based on the reference pathological features and the reference blood clinical indicators;

[0006] Obtaining real-time pathological images and blood test reports of a target patient, and determining the current lung cancer development stage of the target patient according to the real-time pathological images and the blood test reports through the lung cancer stage evaluation model;

[0007] Determining the risk attribute of the target patient based on the current lung cancer development stage, and determining the blood lymphocyte proportion of the target patient based on the risk attribute;

[0008] Determining the key blood indicators of the target patient according to the blood lymphocyte proportion, and outputting the prognosis risk score and the survival prediction result of the target patient through a preset survival analysis model according to the key blood indicators and the multi-scale morphological features.

[0009] Preferably, the reference pathological features and reference blood clinical indicators of the lung cancer patients at different clinical stages are obtained, and a lung cancer stage evaluation model is constructed based on the reference pathological features and the reference blood clinical indicators, including:

[0010] A plurality of development stages of lung cancer diseases are determined, the differentiation degree of tumor cells at each development stage is determined, and the clinical pathological features and vascular infiltration at each development stage are determined according to the differentiation degree of tumor cells;

[0011] The blood metabolomics features at each development stage are determined according to the vascular infiltration, and the reference blood clinical indicators are determined based on the blood metabolomics features and the blood test reports at each development stage;

[0012] A machine learning algorithm and a deep learning architecture and a model training strategy are selected according to the reference blood clinical indicators and the clinical pathological features at each development stage;

[0013] A neural network model is selected based on the deep learning architecture, and the model training is performed based on the machine learning algorithm according to the reference blood clinical indicators and the clinical pathological features at each development stage to generate the lung cancer stage evaluation model.

[0014] Preferably, the real-time pathological images and blood test reports of the target patient are obtained, and the current lung cancer development stage of the target patient is determined according to the real-time pathological images and the blood test reports by the lung cancer stage evaluation model, including:

[0015] The clinical medical history of the target patient is obtained, and the real-time pathological images and blood test reports of the target patient are obtained according to the clinical medical history and the examination item results;

[0016] The whole slice image features of the real-time pathological images are extracted, the current pathological features are determined according to the whole slice image features, and the current blood clinical indicators of the target patient are obtained according to the blood test reports;

[0017] The stage determination condition parameters are generated according to the current blood clinical indicators and the current pathological features, and the stage determination condition parameters are input into the lung cancer stage evaluation model;

[0018] The lung cancer stage evaluation model is matched based on the stage determination condition parameters, and the target development stage with the largest matching degree is selected as the current lung cancer development stage of the target patient.

[0019] Preferably, the risk attribute of the target patient is determined based on the current lung cancer development stage, and the blood lymphocyte proportion of the target patient is determined based on the risk attribute, including:

[0020] The cancer statistical period of the target patient is determined based on the current lung cancer development stage, and the cancer statistical period includes: early stage, early-middle stage, middle stage, middle-late stage, and late stage.

[0021] determine the risk attribute of the target patient based on the cancer distribution characteristics, the risk attribute including: low risk, medium risk and high risk;

[0022] determine the blood lymphocyte change trend of the target patient based on the risk attribute, and determine the correlation index of the blood lymphocyte proportion and the cancer based on the blood lymphocyte change trend;

[0023] determine the blood lymphocyte proportion of the target patient based on the correlation index of the blood lymphocyte proportion and the cancer and the cancer cell statistical parameters.

[0024] Preferably, the key blood indicators of the target patient are determined based on the blood lymphocyte proportion, and the prognosis risk score and the survival period prediction result of the target patient are output by a preset survival analysis model based on the key blood indicators and the multi-scale morphological characteristics, including:

[0025] The biochemical indicator data of the blood sample of the target patient is determined based on the blood lymphocyte proportion, the biochemical indicator data is standardized and the key blood indicators are screened;

[0026] The multi-scale morphological characteristics are extracted from the real-time pathological image, the multi-scale morphological characteristics and the key blood indicators are fused to generate a joint prognosis feature vector;

[0027] The positive correlation between the lymphocytes and the progression-free survival period of the target patient is determined by analyzing the joint prognosis feature vector based on the preset survival analysis model;

[0028] The prognosis risk score and the survival period prediction result of the target patient are determined based on the positive correlation between the lymphocytes and the progression-free survival period.

[0029] A lung cancer patient prognosis evaluation system combining pathological images and blood indicators, the system comprising:

[0030] A construction module for acquiring reference pathological characteristics and reference blood clinical indicators of lung cancer patients at different clinical stages, and constructing a lung cancer stage evaluation model based on the reference pathological characteristics and the reference blood clinical indicators;

[0031] A first determination module for acquiring real-time pathological images and blood test reports of a target patient, and determining the current lung cancer development stage of the target patient based on the real-time pathological images and the blood test reports through the lung cancer stage evaluation model;

[0032] A second determination module for determining the risk attribute of the target patient based on the current lung cancer development stage, and determining the blood lymphocyte proportion of the target patient based on the risk attribute;

[0033] The output module is configured to determine a key blood index of the target patient according to the blood lymphocyte ratio, and output a prognosis risk score and a survival period prediction result of the target patient through a preset survival analysis model according to the key blood index and the multi-scale morphological feature.

[0034] Preferably, the construction module comprises:

[0035] The first determination sub-module is configured to determine a plurality of development stages of the lung cancer disease, determine a degree of differentiation of tumor cells in each development stage, and determine a clinical pathological feature and a blood vessel infiltration condition in each development stage according to the degree of differentiation of tumor cells.

[0036] The second determination sub-module is configured to determine a blood metabolomics feature in each development stage according to the blood vessel infiltration condition, and determine a reference blood clinical index based on the blood metabolomics feature and a blood test report in each development stage.

[0037] The first selection sub-module is configured to select a machine learning algorithm and a deep learning architecture and a model training strategy according to the reference blood clinical index and the clinical pathological feature in each development stage.

[0038] The first generation sub-module is configured to select a neural network model based on the deep learning architecture, and train a model based on the machine learning algorithm according to the reference blood clinical index and the clinical pathological feature in each development stage to generate a lung cancer stage evaluation model.

[0039] Preferably, the first determination module comprises:

[0040] The acquisition sub-module is configured to acquire a clinical medical record of the target patient, and acquire a real-time pathological image and a blood test report of the target patient according to the clinical medical record and an examination item result.

[0041] The third determination sub-module is configured to extract a whole slice image feature of the real-time pathological image, determine a current pathological feature according to the whole slice image feature, and acquire a current blood clinical index of the target patient according to the blood test report.

[0042] The second generation sub-module is configured to generate a stage determination condition parameter according to the current blood clinical index and the current pathological feature, and input the stage determination condition parameter into the lung cancer stage evaluation model.

[0043] The second selection sub-module is configured to match the stage determination condition parameter based on the lung cancer stage evaluation model, and select a target development stage with a maximum matching degree as a current lung cancer development stage of the target patient.

[0044] Preferably, the second determination module comprises:

[0045] A fourth determining sub-module is configured to determine a cancer development stage of the target patient based on the current lung cancer development stage, and the cancer development stage includes a pre-stage, a pre-middle stage, a middle stage, a post-middle stage, and a late stage.

[0046] A fifth determining sub-module is configured to determine a cancer distribution characteristic of lymphocytes according to the cancer development stage, and determine a risk attribute of the target patient based on the cancer distribution characteristic, and the risk attribute includes a low risk, a middle risk, and a high risk.

[0047] A sixth determining sub-module is configured to determine a blood lymphocyte change trend of the target patient based on the risk attribute, and determine a correlation index between the blood lymphocyte proportion and the cancer development according to the blood lymphocyte change trend.

[0048] A seventh determining sub-module is configured to determine the blood lymphocyte proportion of the target patient according to the correlation index between the blood lymphocyte proportion and the cancer development and the cancer cell statistical parameter.

[0049] Preferably, the output module comprises:

[0050] A screening sub-module is configured to determine a biochemical index data of the blood sample of the target patient according to the blood lymphocyte proportion, and perform standardization processing on the biochemical index data and screen a key blood index.

[0051] A third generating sub-module is configured to extract a multi-scale morphological feature according to the real-time pathological image, perform feature fusion on the multi-scale morphological feature and the key blood index, and generate a joint prognosis feature vector.

[0052] An analysis sub-module is configured to determine a positive correlation between the lymphocytes and the progression-free survival period of the target patient based on the joint prognosis feature vector by using a preset survival analysis model.

[0053] An eighth determining sub-module is configured to determine a prognosis risk score and a survival period prediction result of the target patient according to the positive correlation between the lymphocytes and the progression-free survival period.

[0054] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims.

[0055] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0056] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application and explain the technical solutions of the present application, and do not constitute a limitation on the present application.

[0057] Figure 1 A workflow diagram of a lung cancer patient prognosis evaluation method combining pathological images and blood indicators provided by the present application;

[0058] Figure 2 Another workflow diagram of a lung cancer patient prognosis evaluation method combining pathological images and blood indicators provided by the present application;

[0059] Figure 3 A structural schematic diagram of a lung cancer patient prognosis evaluation system combining pathological images and blood indicators provided by the present application;

[0060] Figure 4 A structural schematic diagram of a lung cancer patient prognosis evaluation structure provided by the present application. DETAILED DESCRIPTION

[0061] The exemplary embodiments will be described in detail herein with reference to the drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0062] At present, lung cancer is one of the malignant tumors with the fastest growth of incidence and mortality, and the greatest threat to human health and life. In the past 50 years, many countries have reported that the incidence and mortality of lung cancer have increased significantly. The incidence and mortality of lung cancer in men account for the first place among all malignant tumors, and the incidence in women accounts for the second place, and the mortality accounts for the second place. Therefore, the prognosis evaluation of lung cancer is crucial for treatment decision and survival prediction. At present, the clinic mainly relies on the biopsy sample of the physician based on the unit time point for static evaluation, which has the following problems: the evaluation process mainly depends on artificial experience, it is difficult to capture the dynamic evolution and treatment response of the tumor, the time sequence change information of the patient state in the treatment process is not integrated, so as to ensure the accuracy and reliability of the prognosis survival prediction, and reduce the practicability and stability. In order to solve the above problems, the present embodiment discloses a precise prediction and evaluation method for prognosis survival of patients based on correlation analysis of pathological images and blood clinical indicators.

[0063] A lung cancer patient prognosis evaluation method combining pathological images and blood indicators, as shown in Figure 1 includes the following steps:

[0064] Step S101, acquiring reference pathological features and reference blood clinical indicators of lung cancer patients at different clinical stages, and constructing a lung cancer stage evaluation model based on the reference pathological features and the reference blood clinical indicators;

[0065] In step S102, real-time pathological images and blood test reports of the target patient are acquired, and a current lung cancer development stage of the target patient is determined according to the real-time pathological images and the blood test reports by using a lung cancer stage evaluation model.

[0066] In step S103, a risk attribute of the target patient is determined based on the current lung cancer development stage, and a blood lymphocyte ratio of the target patient is determined based on the risk attribute.

[0067] In step S104, a key blood index of the target patient is determined according to the blood lymphocyte ratio, and a prognosis risk score and a survival period prediction result of the target patient are output by using a preset survival analysis model according to the key blood index and multi-scale morphological features.

[0068] The working principle of the above technical solution is as follows: reference pathological features and reference blood clinical indexes of lung cancer patients at different clinical stages are acquired, a lung cancer stage evaluation model is constructed based on the reference pathological features and the reference blood clinical indexes, real-time pathological images and blood test reports of a target patient are acquired, a current lung cancer development stage of the target patient is determined according to the real-time pathological images and the blood test reports by using the lung cancer stage evaluation model, a risk attribute of the target patient is determined based on the current lung cancer development stage, a blood lymphocyte ratio of the target patient is determined based on the risk attribute, a key blood index of the target patient is determined according to the blood lymphocyte ratio, and a prognosis risk score and a survival period prediction result of the target patient are output by using a preset survival analysis model according to the key blood index and multi-scale morphological features.

[0069] The above technical solution has the following beneficial effects: by constructing the lung cancer stage evaluation model, the development stage of lung cancer of a patient can be accurately determined according to real-time pathological images and blood test reports of the patient, and the survival period and the risk score of the target patient can be accurately predicted and evaluated according to the dynamic cancerization evolution of lymphocytes driven by tumor development combined with the real-time clinical state of the patient, so as to ensure the objectivity, high precision and stability of the prediction result, eliminate the interference of human subjective factors, improve the reliability, and solve the problems of the prior art, such as the evaluation process mainly relying on artificial experience, being difficult to capture the dynamic evolution and treatment response of the tumor, not integrating the time sequence change information of the patient state in the treatment process, and thus being unable to ensure the accuracy and reliability of the prognosis survival period prediction, and reducing the practicality and stability.

[0070] In one embodiment, the acquisition of the reference pathological features and the reference blood clinical indexes of the lung cancer patients at different clinical stages and the construction of the lung cancer stage evaluation model based on the reference pathological features and the reference blood clinical indexes include:

[0071] determine a plurality of development stages of lung cancer diseases, determine a degree of differentiation of tumor cells under each development stage, determine clinical pathological features and vascular infiltration under each development stage according to the degree of differentiation of tumor cells;

[0072] determine a blood metabolomics feature under each development stage according to the vascular infiltration, and determine a reference blood clinical index based on the blood metabolomics feature and a blood test report under each development stage;

[0073] select a machine learning algorithm and a deep learning architecture and a model training strategy according to the reference blood clinical index and the clinical pathological features under each development stage;

[0074] select a neural network model based on the deep learning architecture, and train a model based on the machine learning algorithm according to the reference blood clinical index and the clinical pathological features under each development stage to generate a lung cancer stage evaluation model through the model training strategy.

[0075] In one embodiment, the real-time pathological image and the blood test report of the target patient are obtained, and the current lung cancer development stage of the target patient is determined according to the real-time pathological image and the blood test report through the lung cancer stage evaluation model, including:

[0076] obtain the clinical medical record of the target patient, and obtain the real-time pathological image and the blood test report of the target patient according to the clinical medical record and the examination item results;

[0077] extract full-slice image features of the real-time pathological image, determine current pathological features according to the full-slice image features, and obtain current blood clinical indicators of the target patient according to the blood test report;

[0078] generate stage determination condition parameters according to the current blood clinical indicators and the current pathological features, and input the stage determination condition parameters into the lung cancer stage evaluation model;

[0079] match the stage determination condition parameters based on the lung cancer stage evaluation model, and select a target development stage with the largest matching degree as the current lung cancer development stage of the target patient.

[0080] In one embodiment, the risk attribute of the target patient is determined based on the current lung cancer development stage, and the blood lymphocyte ratio of the target patient is determined based on the risk attribute, including:

[0081] determine a cancer statistical period of the target patient based on the current lung cancer development stage, the cancer statistical period including: early stage, early-middle stage, middle stage, middle-late stage, and late stage;

[0082] determine a lymphocyte canceration distribution characteristic according to the cancer statistical period, and determine the risk attribute of the target patient based on the canceration distribution characteristic, the risk attribute including: low risk, medium risk, and high risk.

[0083] determine a correlation index of the blood lymphocyte proportion and the cancer based on the blood lymphocyte change trend;

[0084] determine the blood lymphocyte proportion of the target patient according to the correlation index of the blood lymphocyte proportion and the cancer and the statistical parameters of the cancer cells.

[0085] In one embodiment, as shown in Figure 2 determine the blood lymphocyte proportion of the target patient according to the correlation index of the blood lymphocyte proportion and the cancer based on the blood lymphocyte change trend;

[0086] Step S201, determine the biochemical index data of the blood sample of the target patient according to the blood lymphocyte proportion, standardize the biochemical index data and select the key blood index;

[0087] Step S202, extract the multi-scale morphological features from the real-time pathological image, fuse the multi-scale morphological features with the key blood index, and generate a joint prognosis feature vector;

[0088] Step S203, analyze and determine the positive correlation between the lymphocyte and the progression-free survival of the target patient based on the joint prognosis feature vector through the preset survival analysis model;

[0089] Step S204, determine the prognosis risk score and survival prediction result of the target patient according to the positive correlation between the lymphocyte and the progression-free survival.

[0090] In one embodiment, the present embodiment also discloses a lung cancer patient prognosis evaluation system combining pathological images and blood indexes, as shown in Figure 3 The system comprises:

[0091] The construction module 301 is used to acquire reference pathological features and reference blood clinical indexes of lung cancer patients at different clinical stages, and construct a lung cancer stage evaluation model based on the reference pathological features and the reference blood clinical indexes;

[0092] The first determination module 302 is used to acquire real-time pathological images and blood test reports of the target patient, and determine the current lung cancer development stage of the target patient according to the real-time pathological images and the blood test reports through the lung cancer stage evaluation model;

[0093] The second determination module 303 is used to determine the risk attribute of the target patient based on the current lung cancer development stage, and determine the blood lymphocyte proportion of the target patient based on the risk attribute;

[0094] The output module 304 is configured to determine a key blood index of the target patient according to the blood lymphocyte ratio, and output a prognosis risk score and a survival period prediction result of the target patient by a preset survival analysis model according to the key blood index and the multi-scale morphological feature.

[0095] The working principle and beneficial effects of the above technical solutions have been described in the method embodiments, and will not be repeated here.

[0096] In one embodiment, as shown in Figure 4 The construction module 301 includes:

[0097] The first determination submodule 3011 is configured to determine a plurality of development stages of lung cancer diseases, determine a tumor cell differentiation degree at each development stage, and determine a clinical pathological feature and a blood vessel infiltration condition at each development stage according to the tumor cell differentiation degree.

[0098] The second determination submodule 3012 is configured to determine a blood metabolomics feature at each development stage according to the blood vessel infiltration condition, and determine a reference blood clinical index based on the blood metabolomics feature and a blood test report at each development stage.

[0099] The first selection submodule 3013 is configured to select a machine learning algorithm and a deep learning architecture and a model training strategy according to the reference blood clinical index and the clinical pathological feature at each development stage.

[0100] The first generation submodule 3014 is configured to select a neural network model based on the deep learning architecture, and generate a lung cancer stage evaluation model by model training based on the machine learning algorithm according to the reference blood clinical index and the clinical pathological feature at each development stage.

[0101] In one embodiment, the first determination module includes:

[0102] The acquisition submodule is configured to acquire a clinical medical record of the target patient, and acquire a real-time pathological image and a blood test report of the target patient according to the clinical medical record and an examination item result.

[0103] The third determination submodule is configured to extract a whole slice image feature of the real-time pathological image, determine a current pathological feature according to the whole slice image feature, and acquire a current blood clinical index of the target patient according to the blood test report.

[0104] The second generation submodule is configured to generate a stage determination condition parameter according to the current blood clinical index and the current pathological feature, and input the stage determination condition parameter into the lung cancer stage evaluation model.

[0105] The second selection sub-module is configured to select a target development stage with the maximum matching degree as the current lung cancer development stage of the target patient by matching the stage determination condition parameters based on the lung cancer stage evaluation model.

[0106] In an embodiment, the second determination module comprises:

[0107] The fourth determination sub-module is configured to determine a cancer development statistical period of the target patient based on the current lung cancer development stage, wherein the cancer development statistical period comprises: early stage, early-middle stage, middle stage, middle-late stage and late stage.

[0108] The fifth determination sub-module is configured to determine a canceration distribution characteristic of lymphocytes according to the cancer development statistical period, and determine a risk attribute of the target patient based on the canceration distribution characteristic, wherein the risk attribute comprises: low risk, medium risk and high risk.

[0109] The sixth determination sub-module is configured to determine a blood lymphocyte change trend of the target patient based on the risk attribute, and determine a correlation index of the blood lymphocyte proportion and canceration according to the blood lymphocyte change trend.

[0110] The seventh determination sub-module is configured to determine the blood lymphocyte proportion of the target patient according to the correlation index of the blood lymphocyte proportion and canceration and the canceration cell statistical parameter.

[0111] In an embodiment, the output module comprises:

[0112] The screening sub-module is configured to determine biochemical index data of the blood sample of the target patient according to the blood lymphocyte proportion, and perform standardization processing and screening on the biochemical index data to obtain key blood indexes.

[0113] The third generation sub-module is configured to extract multi-scale morphological features from the real-time pathological image, perform feature fusion on the multi-scale morphological features and the key blood indexes, and generate a joint prognosis feature vector.

[0114] The analysis sub-module is configured to determine the positive correlation between the lymphocytes and the progression-free survival period of the target patient by analyzing the joint prognosis feature vector based on a preset survival analysis model.

[0115] The eighth determination sub-module is configured to determine a prognosis risk score and a survival period prediction result of the target patient according to the positive correlation between the lymphocytes and the progression-free survival period.

[0116] Those skilled in the art should understand that the first and second in the present application refer to different application stages.

[0117] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

[0118] It is to be understood that the disclosure is not limited to the precise construction herein described and as shown in the attached drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the disclosure. The scope of the disclosure is limited only by the claims that follow.

Claims

1. A method for prognostic assessment of lung cancer patients combining pathological images and blood indicators, characterized in that, Includes the following steps: We obtained reference pathological features and reference blood clinical indicators of lung cancer patients at different clinical stages, and constructed a lung cancer stage assessment model based on the reference pathological features and reference blood clinical indicators. The system acquires real-time pathological images and blood test reports of the target patient, and uses a lung cancer stage assessment model to determine the current lung cancer development stage of the target patient based on the real-time pathological images and blood test reports. Based on the current stage of lung cancer development, the risk attributes of target patients are determined, and based on the risk attributes, the proportion of blood lymphocytes in target patients is determined. The key blood indicators of the target patient are determined based on the proportion of blood lymphocytes. Based on the key blood indicators and multi-scale morphological characteristics, the prognostic risk score and survival prediction results of the target patient are output through a pre-set survival analysis model.

2. The method for prognostic assessment of lung cancer patients combining pathological images and blood indicators according to claim 1, characterized in that, The process involves obtaining reference pathological features and reference blood clinical indicators for lung cancer patients at different clinical stages, and constructing a lung cancer stage assessment model based on these reference pathological features and blood clinical indicators, including: To identify multiple stages of lung cancer development, determine the degree of tumor cell differentiation at each stage, and determine the clinicopathological features and vascular invasion at each stage based on the degree of tumor cell differentiation. Based on the vascular infiltration, the blood metabolomics characteristics at each stage of development were determined, and reference blood clinical indicators were determined based on the blood metabolomics characteristics and blood test reports at each stage of development. Select machine learning algorithms, deep learning architectures, and model training strategies based on reference blood clinical indicators and clinicopathological features at each developmental stage; Based on a deep learning architecture, a neural network model is selected. Through a model training strategy, a machine learning algorithm is used to train the model based on reference blood clinical indicators and clinicopathological features at each development stage to generate a lung cancer stage assessment model.

3. The method for prognostic assessment of lung cancer patients combining pathological images and blood indicators according to claim 1, characterized in that, The process involves acquiring real-time pathological images and blood test reports of the target patient, and determining the current lung cancer development stage of the target patient based on the real-time pathological images and blood test reports using a lung cancer stage assessment model. This includes: Obtain the target patient's clinical medical records, and based on the clinical medical records and examination results, obtain the target patient's real-time pathological images and blood test reports; Extract full-slice image features from real-time pathological images, determine the current pathological features based on the full-slice image features, and obtain the current blood clinical indicators of the target patient based on the blood test report; Based on the current blood clinical indicators and current pathological characteristics, generate stage determination condition parameters and input the stage determination condition parameters into the lung cancer stage assessment model; The lung cancer stage assessment model matches the stage determination criteria parameters and selects the target development stage with the highest matching degree as the current lung cancer development stage of the target patient.

4. The method for prognostic assessment of lung cancer patients combining pathological images and blood indicators according to claim 1, characterized in that, The process of determining the risk attributes of target patients based on their current stage of lung cancer development, and determining the proportion of blood lymphocytes in target patients based on their risk attributes, includes: The cancer statistical period for target patients is determined based on the current stage of lung cancer development. The cancer statistical period includes: early stage, early-mid stage, mid stage, mid-late stage, and very late stage. The distribution characteristics of lymphocyte carcinogenesis are determined based on the cancer statistical period, and the risk attributes of target patients are determined based on the carcinogenesis distribution characteristics. The risk attributes include: low risk, medium risk and high risk. Based on risk attributes, determine the trend of blood lymphocyte changes in target patients, and determine the correlation index between blood lymphocyte ratio and carcinogenesis based on the trend of blood lymphocyte changes. The proportion of blood lymphocytes in the target patient was determined based on the correlation index between the proportion of blood lymphocytes and carcinogenesis and statistical parameters of carcinogenic cells.

5. The method for prognostic assessment of lung cancer patients combining pathological images and blood indicators according to claim 1, characterized in that, The process involves determining key blood indicators for target patients based on blood lymphocyte ratios, and outputting prognostic risk scores and survival predictions for target patients using a pre-defined survival analysis model based on these key blood indicators and multi-scale morphological characteristics. This includes: Based on the proportion of blood lymphocytes, determine the biochemical indicators of blood samples from target patients, standardize the biochemical indicators, and screen key blood indicators. Multi-scale morphological features are extracted from real-time pathological images, and these features are fused with key blood indicators to generate a joint prognostic feature vector. The positive correlation between lymphocyte count and progression-free survival in target patients was determined by analyzing the combined prognostic feature vectors using a pre-defined survival analysis model. The prognostic risk score and survival prediction results for target patients were determined based on the positive correlation between lymphocytes and progression-free survival.

6. A prognostic assessment system for lung cancer patients combining pathological images and blood indicators, characterized in that, The system includes: A module is built to obtain reference pathological features and reference blood clinical indicators of lung cancer patients at different clinical stages, and to build a lung cancer stage assessment model based on the reference pathological features and reference blood clinical indicators. The first determination module is used to acquire real-time pathological images and blood test reports of the target patient, and to determine the current lung cancer development stage of the target patient based on the real-time pathological images and blood test reports through the lung cancer stage assessment model; The second determination module is used to determine the risk attributes of the target patient based on the current stage of lung cancer development, and to determine the proportion of blood lymphocytes in the target patient based on the risk attributes. The output module is used to determine the key blood indicators of the target patient based on the proportion of blood lymphocytes, and output the prognostic risk score and survival prediction results of the target patient through a preset survival analysis model based on the key blood indicators and multi-scale morphological characteristics.

7. The lung cancer patient prognostic assessment system combining pathological images and blood indicators according to claim 6, characterized in that, The building module includes: The first determination submodule is used to determine multiple developmental stages of lung cancer, determine the degree of tumor cell differentiation in each developmental stage, and determine the clinicopathological features and vascular invasion in each developmental stage based on the degree of tumor cell differentiation. The second determination submodule is used to determine the blood metabolomics characteristics at each stage of development based on the vascular infiltration, and to determine reference blood clinical indicators based on the blood metabolomics characteristics and blood test reports at each stage of development. The first selection submodule is used to select machine learning algorithms, deep learning architectures, and model training strategies based on reference blood clinical indicators and clinicopathological features at each development stage. The first generation submodule is used to select a neural network model based on a deep learning architecture, and generate a lung cancer stage assessment model by training the model using a model training strategy based on machine learning algorithms according to reference blood clinical indicators and clinicopathological features at each development stage.

8. The lung cancer patient prognostic assessment system combining pathological images and blood indicators according to claim 6, characterized in that, The first determining module includes: The acquisition submodule is used to acquire the clinical medical records of the target patient and to acquire the real-time pathological images and blood test reports of the target patient based on the clinical medical records and test results. The third determination submodule is used to extract the full-slice image features of real-time pathological images, determine the current pathological features based on the full-slice image features, and obtain the current blood clinical indicators of the target patient based on the blood test report; The second generation submodule is used to generate stage determination condition parameters based on the current blood clinical indicators and current pathological characteristics, and input the stage determination condition parameters into the lung cancer stage assessment model. The second selection submodule is used to match the lung cancer stage assessment model based on stage determination condition parameters and select the target development stage with the highest matching degree as the current lung cancer development stage of the target patient.

9. The prognostic assessment system for lung cancer patients combining pathological images and blood indicators according to claim 6, characterized in that, The second determining module includes: The fourth determination submodule is used to determine the cancer statistical period of the target patient based on the current stage of lung cancer development. The cancer statistical period includes: early stage, early-mid stage, mid stage, mid-late stage, and very late stage. The fifth determination submodule is used to determine the distribution characteristics of lymphocyte carcinogenesis based on the cancer statistical period, and to determine the risk attributes of the target patient based on the carcinogenesis distribution characteristics. The risk attributes include: low risk, medium risk and high risk. The sixth submodule is used to determine the trend of blood lymphocyte changes in target patients based on risk attributes, and to determine the correlation index between blood lymphocyte ratio and cancer based on the trend of blood lymphocyte changes. The seventh determination submodule is used to determine the proportion of blood lymphocytes in the target patient based on the correlation index between the proportion of blood lymphocytes and carcinogenesis and the statistical parameters of carcinogenic cells.

10. The lung cancer patient prognostic assessment system combining pathological images and blood indicators according to claim 6, characterized in that, The output module includes: The screening submodule is used to determine the biochemical indicators of the blood samples of the target patient based on the proportion of blood lymphocytes, standardize the biochemical indicator data, and screen key blood indicators. The third generation submodule is used to extract multi-scale morphological features from real-time pathological images, fuse the multi-scale morphological features with key blood indicators, and generate a joint prognostic feature vector. The analysis submodule is used to determine the positive correlation between lymphocytes and progression-free survival in target patients by analyzing the combined prognostic feature vectors based on a preset survival analysis model. The eighth determination submodule is used to determine the prognostic risk score and survival prediction results of the target patient based on the positive correlation between lymphocytes and progression-free survival.

Images