A prediction method and system for the pathological state of lymph nodes after immunotherapy

Abstract

The present invention provides a method and system for predicting the pathological status of lymph nodes after immunotherapy, which are applied to the field of data processing technology. This application obtains the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, a training sample set, and a preset prediction model for the pathological remission rate of metastatic lymph nodes; processes the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; processes the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor; processes the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; processes the lymph node residue information of the target user to generate the adjuvant therapy intervention information of the target user.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and system for predicting the pathological status of lymph nodes after immunotherapy. Background Art

[0002] In recent years, neoadjuvant immunotherapy has been applied in the treatment of non-small cell lung cancer, which has improved the pathological remission rate of patients and significantly improved their survival status. Although neoadjuvant immunotherapy has brought certain positive effects, more than half of the patients still fail to obtain sufficient benefits from the treatment. Therefore, there is an urgent need to screen out patients with poor treatment effects.

[0003] Currently, neoadjuvant treatment research mainly focuses on the evaluation of primary tumors, while lymph nodes have not received enough attention. Clinical practice has shown that in addition to primary tumors, the pathological status of lymph nodes is also closely related to the survival and prognosis of patients. On the one hand, the independent evaluation of lymph nodes has not become a clinical standard; on the other hand, existing evaluation methods are ineffective in predicting the pathological remission rate of lymph nodes, and there is no effective method for predicting the pathological remission status of lymph nodes.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present application is to provide a method and system for predicting the pathological status of lymph nodes after immunotherapy, which at least to some extent overcomes the problems existing in the prior art. By obtaining the clinical, imaging, and metabolic parameter information of the target user, as well as the training sample set and the preset model, the preset model is processed with the training sample set, and through operations such as data feature acquisition, sampling, grouping, and missing value imputation, a prediction model for the pathological remission rate of target metastatic lymph nodes is generated. Then, features are extracted from the multi-source parameter information and regression analysis is performed to obtain risk prediction factors including the diameter of the primary tumor after treatment, etc. The risk prediction factors are input into the target model, and the predicted value of the pathological remission status of the lymph nodes is calculated using a specific formula, and then the residual information of the lymph nodes is obtained. Finally, the residual information of the lymph nodes is quantitatively analyzed, combined with the changes in imaging and metabolic parameter information within the preset time period, as well as clinical parameters, to generate residual lymph node association information, and based on this, auxiliary treatment intervention information is formulated.

[0006] Other features and advantages of the present application will become apparent through the following detailed description, or will be learned in part through the practice of the present invention.

[0007] According to one aspect of the present application, there is provided a method for predicting the pathological status of lymph nodes after immunotherapy, including: obtaining the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, a training sample set, and a preset prediction model for the pathological remission rate of metastatic lymph nodes, wherein the target user is a user with lymphatic metastatic lung cancer, and the training sample set is patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment; processing the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; processing the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor; processing the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; and processing the lymph node residue information of the target user to generate the adjuvant treatment intervention information of the target user.

[0008] According to another aspect of the present application, there is provided a device for predicting the pathological status of lymph nodes after immunotherapy, characterized by including: an acquisition module, configured to obtain the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, a training sample set, and a preset prediction model for the pathological remission rate of metastatic lymph nodes, wherein the target user is a user with lymphatic metastatic lung cancer, and the training sample set is patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment; a processing module, configured to process the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; process the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor; process the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; and process the lymph node residue information of the target user to generate the adjuvant treatment intervention information of the target user.

[0009] According to still another aspect of the present application, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a second processor, the method for predicting the pathological status of lymph nodes after immunotherapy as described above is implemented.

[0010] A prediction method and system for the pathological status of lymph nodes after immunotherapy provided by the present application. The present application obtains the clinical, imaging, and metabolic parameter information of the target user, as well as the training sample set and the preset model. The preset model is processed with the training sample set, and through operations such as data feature acquisition, sampling, grouping, and missing value imputation, a prediction model for the pathological remission rate of the target metastatic lymph nodes is generated. Then, features are extracted from the multi-source parameter information and regression analysis is performed to obtain risk prediction factors including the diameter of the primary tumor after treatment, etc. The risk prediction factors are input into the target model, and the predicted value of the pathological remission status of the lymph nodes is calculated using a specific formula, and then the residual information of the lymph nodes is obtained. Finally, the residual information of the lymph nodes is quantitatively analyzed, combined with the change information of the imaging and metabolic parameters within the preset time period, and the clinical parameters, to generate the residual information associated with the lymph nodes, and based on this, the auxiliary treatment intervention information is formulated.

[0011] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Brief Description of the Drawings

[0012] Figure 1 The flowchart showing a prediction method for the pathological status of lymph nodes after immunotherapy provided by an embodiment of the present application;

[0013] Figure 2 The structural schematic diagram showing a prediction device for the pathological status of lymph nodes after immunotherapy provided by an embodiment of the present application. Detailed Embodiments

[0014] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0015] The following combines Figure 1 to describe the prediction method for the pathological status of lymph nodes after immunotherapy according to the exemplary embodiments of the present application. In one embodiment, the present application also proposes a prediction method and system for the pathological status of lymph nodes after immunotherapy. Figure 1 Schematically shows the flowchart of a prediction method for the pathological status of lymph nodes after immunotherapy according to an embodiment of the present application. As Figure 1 shown, this method is applied to the server and includes:

[0016] S101, obtaining the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, the training sample set, and the preset prediction model for the pathological remission rate of metastatic lymph nodes.

[0017] In one implementation, for target users with lymphatic metastatic lung cancer who have completed neoadjuvant immunotherapy, the clinical parameter information covers multiple key aspects. Age is an important factor. There are differences in the physical functions, treatment tolerances, and tumor development speeds of patients in different age groups. For example, elderly patients have a lower tolerance for high-intensity treatments. Gender also affects tumor development and the response to treatment. Clinical studies have found that there are certain differences in the incidence, pathological types, and treatment effects of lung cancer between men and women. Smoking history is closely related to the occurrence and development of lung cancer. The tumor cells of long-term smokers have unique biological characteristics, which affect the treatment effect. The clinical stage of the tumor directly reflects the severity and spread range of the tumor and is an important basis for formulating treatment strategies. As a key biomarker, the PD-L1 expression level is closely related to the efficacy of immunotherapy. Patients with high expression are more sensitive to immunotherapy. The pathological type cannot be ignored either. Different pathological types of lung cancer (such as adenocarcinoma, squamous cell carcinoma, etc.) vary in biological behavior and treatment sensitivity.

[0018] The imaging parameter information is mainly obtained through chest CT or PET / CT scans. The diameter of the primary tumor before treatment, the diameter of the primary tumor after treatment, and the change rate between the two can directly reflect the size change of the tumor during the treatment process. If the diameter of the primary tumor significantly shrinks after treatment and the change rate is large, it usually indicates a good treatment effect. The diameter of the lymph nodes before treatment, the diameter of the lymph nodes after treatment, and the change rate of the lymph node diameter before and after treatment are crucial for judging lymph node metastasis and treatment effect. The change in the lymph node diameter implies the proliferation or regression of tumor cells in the lymph nodes and is an important indicator for evaluating treatment effect and disease progression. The metabolic parameter information mainly includes the metabolic value of the primary tumor before treatment, the metabolic value of the primary tumor after treatment, and the change rate of the metabolic value, as well as the metabolic value of the lymph nodes before treatment, the metabolic value of the lymph nodes after treatment, and the change rate of the metabolic value. These parameters are measured by PET / CT scans, and the maximum standardized uptake value (SUVmax) is a commonly used indicator. The change in the metabolic value reflects the change in the metabolic activity of tumor cells. A decrease in the metabolic activity of tumor cells usually means that the treatment effectively inhibits tumor growth; conversely, an increase in metabolic activity indicates tumor progression or poor treatment effect.

[0019] The training sample set selects patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment. The treatment process and outcome data of these patients are valuable resources for model training. The sample set contains various clinical characteristics, imaging, and metabolic parameter information before treatment, as well as the results of lymph node pathological status after treatment. By learning from a large number of such samples, the model can uncover the potential relationships between different parameters and the lymph node pathological status. For example, by analyzing the lymph node pathological remission of patients under different combinations of clinical characteristics, imaging, and metabolic parameters, it provides a basis for predicting the lymph node pathological status of new target users. The diversity and representativeness of the sample set are crucial. Covering patients of different ages, genders, pathological types, and treatment regimens can improve the generalization ability of the model, enabling it to have better prediction performance when facing various actual cases.

[0020] The preset prediction model for lymph node pathological remission rate uses Lasso regression. Its core lies in analyzing a large amount of relevant data, screening key features, and constructing a prediction formula to effectively predict the lymph node pathological remission rate. Lasso regression (Least Absolute Shrinkage and Selection Operator regression) is an extension of linear regression. It introduces an L1 regularization term on the basis of ordinary linear regression. This feature enables Lasso regression not only to perform variable selection, screening out variables that have a significant impact on the dependent variable (lymph node pathological remission status) from numerous independent variables (such as clinical parameters, imaging parameters, metabolic parameters, etc.), but also to compress the regression coefficients, effectively avoiding model overfitting and improving the generalization ability of the model. When predicting the lymph node pathological remission rate of metastatic lymph nodes, Lasso regression can accurately identify key prediction factors in the case of high-dimensional data (i.e., data containing a large number of features) and construct a concise and effective prediction model.

[0021] When constructing the preset prediction model for lymph node pathological remission rate based on Lasso regression, it can be generally divided into a data preprocessing layer, a model training layer, and a model output layer. The data preprocessing layer is responsible for cleaning, standardizing, and handling missing values for the clinical parameter information, imaging parameter information, metabolic parameter information of the target users collected, as well as the training sample set, ensuring the quality and consistency of the input data, and providing reliable data for subsequent model training. The model training layer is the core layer. It uses the Lasso regression algorithm to learn the preprocessed data, continuously adjusts the regression coefficients, and searches for the optimal solution to enable the model to accurately capture the relationship between each parameter and the lymph node pathological remission status. The model output layer calculates the new input target user data according to the trained model and outputs the prediction result of the lymph node pathological remission status.

[0022] Structurally, the model based on Lasso regression mainly consists of a regression coefficient vector and a regularization parameter. The regression coefficient vector determines the weight of each independent variable in the prediction. For example, when predicting the pathological remission rate of lymph nodes, independent variables such as the PT diameter after treatment, the reduction rate of PT diameter, PT-ΔSUVmax, and LN-ΔSUVmax all have corresponding regression coefficients, and these coefficients reflect the contribution degree of each independent variable to the prediction result. In addition to the above-mentioned regression coefficients and regularization parameters, there are also some important parameters related to model construction and training. In the data preprocessing stage, the standardization process involves calculating the mean and standard deviation of the data, which are used to normalize the data so that different features have the same scale and avoid affecting the model training effect due to excessive differences in feature scales. During the model training process, the number of iterations determines the number of calculations when the algorithm searches for the optimal solution. An appropriate number of iterations can ensure that the model converges to a better result. If the number of iterations is too small, the model does not fully learn the data features, resulting in inaccurate predictions; if the number of iterations is too large, it will increase the calculation time and overfitting will occur. In addition, the learning rate controls the step size of the regression coefficient update in each iteration, and an appropriate learning rate can accelerate the model convergence speed and improve the training efficiency.

[0023] S102, process the preset metastatic lymph node pathological remission rate prediction model based on the training sample set to generate a target metastatic lymph node pathological remission rate prediction model.

[0024] In one implementation, obtain any number of data features in the training sample set, and select some data features for analysis from the training sample set of these 1000 patients. For example, select 20 data features such as age, gender, smoking history, tumor clinical stage, PD-L1 expression level, pathological type, primary tumor diameter before treatment, primary tumor diameter after treatment, change rate of primary tumor diameter before and after treatment, lymph node diameter before treatment, lymph node diameter after treatment, change rate of lymph node diameter before and after treatment, primary tumor metabolic value before treatment, primary tumor metabolic value after treatment, change rate of primary tumor metabolic value before and after treatment, lymph node metabolic value before treatment, lymph node metabolic value after treatment, and change rate of lymph node metabolic value before and after treatment. These features cover the basic clinical information, imaging information, and metabolic information of the patients, and are of great significance for predicting the pathological remission rate of lymph nodes.

[0025] Generate a sampling ratio based on the quantity of each data feature in the training sample set, sample the training sample set according to the sampling ratio, and generate a preset number of sampled features. Calculate the proportion of the quantity of each of the selected 20 data features to the total number of features to obtain the sampling ratio. Suppose among these 20 features, the data integrity of the age feature is relatively high, with 950 valid data. Then the sampling ratio of the age feature is 950÷1000 = 0.95; while for a certain metabolic value feature, due to detection technology limitations, there are only 800 valid data, and its sampling ratio is 800÷1000 = 0.8. In this way, a sampling weight is determined for each data feature to ensure that in the subsequent sampling process, different features can be reasonably sampled according to their data richness. Sample the training sample set according to the calculated sampling ratio. Set the preset number to 500 sampled features, and use the method of random sampling combined with weights to extract data from the training sample set. For example, when extracting data about age, according to the sampling ratio of 0.95, it is expected to extract 950×(500÷1000) = 475 data; for the metabolic value feature, according to the sampling ratio of 0.8, it is expected to extract 800×(500÷1000) = 400 data. After sampling, a subset containing 500 sampled features is obtained. This subset contains partial data of different data features and retains the distribution characteristics of the original data features.

[0026] Process any data feature with each sampled feature to generate a response group and a non - response group. Among them, each data group contains a preset number of data samples, and at least one data sample includes identification information. Select one of the data features, such as the diameter of the primary tumor after treatment, and compare and analyze it with each sampled feature. Set a threshold. Suppose that the diameter of the primary tumor after treatment less than 1 cm is an indicator of better treatment response. According to this criterion, the patient data corresponding to 500 sampled features are divided into two groups. The patient data with the diameter of the primary tumor after treatment less than 1 cm are classified into the response group, which contains 300 data samples; the patient data greater than or equal to 1 cm are classified into the non - response group, which contains 200 data samples. Among these data samples, each sample contains various information of the patient, such as age, gender, etc., and at least one data sample contains information that can identify whether the patient has risk factors affecting lymphatic metastatic lung cancer, such as specific gene markers or special clinical symptoms, etc.

[0027] Missing value imputation was performed on the responder group and the poor responder group respectively to generate missing value imputation prediction information. Based on the data samples in the responder group and the poor responder group, a preset prediction model for the pathological remission rate of metastatic lymph nodes was trained to generate a trained prediction model for the pathological remission rate of metastatic lymph nodes. By checking the data of the responder group and the poor responder group, it was found that there were some missing data. For the missing age data in the responder group, the multiple imputation method was used for processing. Using the relevant information such as the age, gender, and tumor stage of other patients in the group, a regression model was constructed to predict the missing age value. For example, through analysis, it was found that there was a certain correlation between age and tumor stage, and a regression equation was established using this relationship to predict the missing age data. Similarly, similar processing was also performed on the missing data in the poor responder group. After processing, more complete data for the responder group and the poor responder group were obtained, and these supplementary data were the missing value imputation prediction information. The data of the responder group and the poor responder group after missing value imputation were input into a preset prediction model for the pathological remission rate of metastatic lymph nodes based on Lasso regression. The model began to learn the relationship between the features in the data and the pathological remission rate of lymph nodes, and adjusted the regression coefficients in the model, such as the coefficients of parameters such as the PT diameter after treatment and the PT diameter reduction rate. After multiple iterative trainings, the model gradually fit the data to obtain a trained prediction model for the pathological remission rate of metastatic lymph nodes. During the training process, the model was continuously optimized to improve the prediction accuracy of the pathological remission rate of lymph nodes.

[0028] Process the trained prediction model for the pathological remission rate of metastatic lymph nodes to generate a verification result. If the data sample containing identification information in the verification result represents a risk factor affecting lymphatic metastatic lung cancer, then use the trained prediction model for the pathological remission rate of metastatic lymph nodes as the target prediction model for the pathological remission rate of metastatic lymph nodes. Use the data of another 500 patients in the training sample set who did not participate in the training to verify the trained model. Input the relevant data of these patients into the model, and the model outputs the predicted results of the pathological remission rate of the lymph nodes of these patients. Then, compare the predicted results with the actual lymph node pathological examination results of these patients, and calculate indicators such as the prediction accuracy, specificity, and sensitivity of the model. For example, if the model predicts the pathological remission status of 400 patients, and 300 of them are predicted correctly, then the prediction accuracy is 300÷400 = 75%. In the verification result, check the data sample containing identification information. Assume that the identification information is a specific gene marker. If the data sample carrying this gene marker shows that patients with this marker have a higher risk of lymph node metastasis in the model prediction and is consistent with the actual clinical situation, that is, the pathological remission of the lymph nodes of these patients is indeed poor, then it indicates that this trained model can effectively identify the risk factors affecting lymphatic metastatic lung cancer. At this time, use the trained prediction model for the pathological remission rate of metastatic lymph nodes as the target prediction model for the pathological remission rate of metastatic lymph nodes for subsequent prediction of the pathological remission rate of the lymph nodes of the target user.

[0029] In another implementation, the missing sample information of the responder group and the non-responder group is obtained respectively, and the responder group and the non-responder group are processed based on the missing sample information of the responder group and the non-responder group respectively to generate predicted values for the missing samples. Data missing may exist in both the responder group and the non-responder group. For example, in the responder group, it is found that the PD-L1 expression level data of some patients is missing, involving a total of 50 patients; in the non-responder group, the pre-treatment lymph node metabolic value is missing, involving 30 patients. These missing data may affect the accuracy of model training, so further processing is required. For the missing PD-L1 expression level data in the responder group, we use a machine learning-based method to predict the missing values. Using other complete clinical parameters (such as age, gender, smoking history, tumor clinical stage, pathological type), imaging parameters (pre-treatment primary tumor diameter, post-treatment primary tumor diameter, change rate of primary tumor diameter before and after treatment, pre-treatment lymph node diameter, post-treatment lymph node diameter, change rate of lymph node diameter before and after treatment) and metabolic parameters (pre-treatment primary tumor metabolic value, post-treatment primary tumor metabolic value, change rate of primary tumor metabolic value before and after treatment, post-treatment lymph node metabolic value, change rate of lymph node metabolic value before and after treatment) in the responder group as features, a random forest algorithm is used to construct a prediction model. After training, the 50 patients with missing PD-L1 expression levels are predicted to obtain predicted values. For example, after model prediction, the predicted PD-L1 expression level of one patient is 50%.

[0030] For the missing pre-treatment lymph node metabolic value in the non-responder group, a prediction model is also constructed based on other complete parameters within the group. Here, a linear regression model is used, with age, tumor clinical stage, pre-treatment primary tumor metabolic value, etc. as independent variables, and 30 patients with missing pre-treatment lymph node metabolic values are predicted. Suppose the predicted pre-treatment lymph node metabolic value of a patient is 3.5 (SUVmax).

[0031] The true values corresponding to the missing samples are obtained, and the predicted values of the missing samples and the true values corresponding to the missing samples are processed to generate missing value imputation prediction information. Taking the processing of the missing PD-L1 expression level in the responder group as an example, assume that the formula parameters obtained through training with a large number of samples are: , regression coefficient related to clinical parameters (taking age as an example, assuming the age range is 1 - 100 years old and has been standardized) , regression coefficient related to imaging parameters (taking the post-treatment primary tumor diameter as an example, assuming the diameter range is 1 - 10 cm and has been standardized) , regression coefficient related to metabolic parameters (taking the post-treatment primary tumor metabolic value as an example, assuming the metabolic value range is 1 - 10 and has been standardized) 。For a patient with a missing PD-L1 expression level, the age standardized value is 0.6, the standardized value of the primary tumor diameter after treatment is 0.4, and the standardized value of the primary tumor metabolic value after treatment is 0.5. The method also includes a calculation formula for obtaining imputation prediction information for missing values, and the calculation formula is: ; where represents the constant term of the formula, represents the regression coefficient related to clinical parameters, represents the value of the kth clinical parameter, represents the regression coefficient related to imaging parameters, represents the value of the th imaging parameter, represents the regression coefficient related to metabolic parameters, represents the value of the th metabolic parameter, and is calculated according to the formula:

[0032] The calculated result is inverse standardized (assuming the inverse standardized range is 0-100%), and the imputation prediction information for the missing value of the PD-L1 expression level of this patient is obtained as 40.5% (the actual inverse standardization calculation is determined according to the standardization method of the original data). In this way, the predicted values and true values of all missing samples are processed to obtain complete imputation prediction information for missing values, which is used for subsequent model training to improve the accuracy and reliability of the model.

[0033] S103. Process the clinical parameter information, imaging parameter information, and metabolic parameter information of the target user to generate a risk prediction factor.

[0034] In one implementation, feature extraction processing is performed on the clinical parameter information of the target user to generate the clinical parameter features of the target user. Among them, the clinical parameter features of the target user include age, gender, smoking history, tumor clinical stage, PD-L1 expression level, and pathological type. Suppose the target user is a 58-year-old male non-small cell lung cancer patient with a 30-year smoking history, a tumor clinical stage of IIIB, a PD-L1 expression level of 50%, and a pathological type of adenocarcinoma. Feature extraction is performed on the clinical parameter information of this target user. Age, as a continuous variable, is directly recorded as 58 years old. The gender is male, and an encoding method can be used. For example, "1" represents male and "0" represents female. The smoking history is 30 years, which can be used as a quantitative index or divided into different levels according to medical research. For example, 0-10 years is a mild smoking history, 11-30 years is a moderate smoking history, and more than 30 years is a heavy smoking history. This patient belongs to the heavy smoking history. The tumor clinical stage IIIB is recorded according to the medical standard staging. The PD-L1 expression level of 50% is a key indicator reflecting the sensitivity of immunotherapy, and this value is directly recorded. The pathological type is adenocarcinoma, which is also recorded according to the medical classification standard. Through these processes, the clinical parameter features of the target user are generated. These features provide basic information for subsequent analysis from aspects such as the patient's basic situation, disease characteristics, and immune-related indicators.

[0035] Feature extraction processing is performed on the imaging parameter information of the target user to generate the imaging features of the target user. Among them, the imaging features of the target user include the diameter of the primary tumor before treatment, the diameter of the primary tumor after treatment, the change rate of the diameter of the primary tumor before and after treatment, the diameter of the lymph nodes before treatment, the diameter of the lymph nodes after treatment, and the change rate of the diameter of the lymph nodes before and after treatment. The patient had chest CT scans before the first treatment and after neoadjuvant chemotherapy respectively. The diameter of the primary tumor before treatment was measured to be 4 cm, and the diameter of the primary tumor after treatment shrank to 2 cm. Then the change rate of the diameter of the primary tumor before and after treatment is (4 - 2) ÷ 4 × 100% = 50%. The diameter of the lymph nodes before treatment was measured to be 1.5 cm, and it shrank to 1 cm after treatment. The change rate of the diameter of the lymph nodes before and after treatment is (1.5 - 1) ÷ 1.5 × 100% ≈ 33.3%. These data intuitively reflect the size changes of the tumor and lymph nodes during the treatment process and are important bases for evaluating the treatment effect and disease progression. By extracting the imaging parameter information, the imaging features of the target user are obtained, and these features provide a visual quantitative basis for judging the status of the tumor and lymph nodes.

[0036] Feature extraction is performed on the metabolic parameter information of the target user to generate the metabolic parameter features of the target user. Among them, the metabolic parameter features of the target user include the pre-treatment primary tumor metabolic value, the post-treatment primary tumor metabolic value, the change rate of the primary tumor metabolic value before and after treatment, the pre-treatment lymph node metabolic value, the post-treatment lymph node metabolic value, and the change rate of the lymph node metabolic value before and after treatment. The metabolic parameter information of the patient is obtained by PET / CT scan. The pre-treatment primary tumor metabolic value (expressed as SUVmax) is 8.0, which drops to 4.0 after treatment. The change rate of the primary tumor metabolic value before and after treatment is (8.0 - 4.0) ÷ 8.0 × 100% = 50%. The pre-treatment lymph node metabolic value (SUVmax) is 6.0, which becomes 3.0 after treatment. The change rate of the lymph node metabolic value before and after treatment is (6.0 - 3.0) ÷ 6.0 × 100% = 50%. The changes in these metabolic parameters reflect the changes in the metabolic activity of tumor cells. A decrease in the metabolic value usually means that the treatment effectively inhibits tumor growth. By extracting the metabolic parameter information, the metabolic parameter features of the target user are generated, providing an important reference for evaluating the biological behavior of tumors.

[0037] Regression analysis is performed on the clinical parameter features of the target user, the imaging features of the target user, and the metabolic parameter features of the target user to generate risk prediction factors. Among them, the risk prediction factors include the post-treatment primary tumor diameter, the reduction rate of the primary tumor diameter, the change amount of the maximum standardized uptake value of the primary tumor, and the change amount of the maximum standardized uptake value of the lymph node. The above-extracted clinical parameter features, imaging features, and metabolic parameter features are integrated for regression analysis. The purpose of regression analysis is to find the potential relationship between these features and the pathological remission status of the lymph nodes. In this process, statistical software or machine learning algorithms are used to perform regression calculations with the post-treatment primary tumor diameter, the reduction rate of the primary tumor diameter, the change amount of the maximum standardized uptake value of the primary tumor (i.e., the SUVmax change amount corresponding to the change rate of the primary tumor metabolic value before and after treatment), and the change amount of the maximum standardized uptake value of the lymph node (i.e., the SUVmax change amount corresponding to the change rate of the lymph node metabolic value before and after treatment) as the dependent variables and other features as the independent variables. For example, through regression analysis, it is found that for every 1 cm increase in the post-treatment primary tumor diameter, the risk of pathological remission of the lymph nodes decreases by a certain proportion; for every 10% increase in the reduction rate of the primary tumor diameter, the possibility of pathological remission of the lymph nodes increases to a certain extent, etc. Through regression analysis, risk prediction factors are finally generated. These factors comprehensively reflect the influence degree of multiple features on the pathological remission status of the lymph nodes, providing key inputs for subsequent evaluation of lymph node residual information using the prediction model.

[0038] S104. Process the risk prediction factors based on the target metastatic lymph node pathological remission rate prediction model to generate the lymph node residual information of the target user.

[0039] In one implementation, a risk prediction factor is processed based on a prediction model for the pathological remission rate of target metastatic lymph nodes to generate a predicted value of the pathological remission state of the lymph nodes. The known risk prediction factors include the diameter of the primary tumor after treatment, the shrinkage rate of the primary tumor diameter, the change in the maximum standardized uptake value of the primary tumor, and the change in the maximum standardized uptake value of the lymph nodes. Suppose that through previous examinations and calculations, the diameter of the primary tumor after treatment of this patient is 2 cm, the shrinkage rate of the primary tumor diameter is 50%, the change in the maximum standardized uptake value of the primary tumor (i.e., the change in SUVmax corresponding to the change rate of the primary tumor metabolic value before and after treatment) is 4.0 (assuming the SUVmax before treatment is 8.0 and after treatment is 4.0), and the change in the maximum standardized uptake value of the lymph nodes (i.e., the change in SUVmax corresponding to the change rate of the lymph node metabolic value before and after treatment) is 3.0 (assuming the SUVmax of the lymph nodes before treatment is 6.0 and after treatment is 3.0).

[0040] According to the formula for calculating the predicted value of the pathological remission state of the lymph nodes: Predicted value of the pathological remission state of the lymph nodes = 0.0253 × diameter of the primary tumor after treatment - 0.0221 × shrinkage rate of the primary tumor diameter - 0.2838 × change in the maximum standardized uptake value of the primary tumor - 0.0359 × change in the maximum standardized uptake value of the lymph nodes + 0.2093. Substituting the patient's data into the formula gives: . This calculation result -1.0 is the predicted value of the pathological remission state of the lymph nodes of this patient obtained based on the prediction model for the pathological remission rate of target metastatic lymph nodes. The meaning of this value is a quantitative evaluation result calculated by the model integrating various risk prediction factors, used to reflect the possibility of pathological remission of the lymph nodes. Here, it is negative, preliminarily indicating that the pathological remission state of the lymph nodes of this patient may not be ideal.

[0041] Process the predicted value of the lymph node pathological remission status to generate the lymph node residual information of the target user. After obtaining the predicted value of the lymph node pathological remission status, it is necessary to further process it to generate the lymph node residual information of the target user. A threshold can be set according to clinical experience and a large amount of research data to classify the lymph node residual situation. For example, when the predicted value is greater than 0, it is considered that the lymph node residual is less and the pathological remission status is better; when the predicted value is less than or equal to 0, it is considered that there is a certain degree of lymph node residual and the pathological remission status is poor. Since the predicted value of this patient is -1.0, which is less than 0, it can be judged that there is a certain degree of tumor residual in the lymph nodes of this patient. Specifically, it can be further described as follows: According to the model prediction, after receiving neoadjuvant immunotherapy, there is a high possibility of tumor cell residual in the lymph nodes of this 58-year-old male non-small cell lung cancer patient, and the pathological remission status of his lymph nodes is not ideal, which may affect the subsequent treatment and prognosis of the patient. Doctors can consider whether to adjust the treatment plan based on this result and the patient's other clinical information, such as strengthening subsequent adjuvant treatment measures, etc.

[0042] S105. Process the lymph node residual information of the target user to generate the adjuvant treatment intervention information of the target user.

[0043] In one implementation, a quantitative analysis process is performed on the lymph node residual information of the target user to generate the lymph node residual degree information. Assume that through the previous calculation, the predicted value of the lymph node pathological remission status of this patient is -1.0, and it is judged that there is lymph node residual. A quantitative standard can be set according to clinical experience and a large amount of research data. For example, when the predicted value is -1.5 and below, it is highly residual, between -1.5 and -0.5 is moderately residual, and above -0.5 is lowly residual. The predicted value of this patient, -1.0, is between -1.5 and -0.5, so it is determined that the degree of his lymph node residual is moderate.

[0044] Obtain the imaging parameter information and metabolic parameter information of the target user within a preset time period, process the imaging parameter information and metabolic parameter information of the target user within the preset time period, and generate imaging feature change information and metabolic parameter feature change information. At 1 month and 3 months after the patient completes neoadjuvant immunotherapy, chest CT and PET / CT scans are performed on the patient respectively. At 1 month, the diameter of the primary tumor after treatment is measured to be 2 cm, and the diameter of the lymph node after treatment is 1 cm; at 3 months, the diameter of the primary tumor shrinks to 1.5 cm, and the diameter of the lymph node shrinks to 0.8 cm. Metabolic parameters are obtained from the PET / CT scan. At 1 month, the metabolic value (SUVmax) of the primary tumor is 3.5, and the metabolic value of the lymph node is 2.5; at 3 months, the metabolic value of the primary tumor drops to 2.8, and the metabolic value of the lymph node drops to 2.0. Calculate the imaging feature change information. The change amount of the diameter of the primary tumor within 1 - 3 months is 1.5 - 2 = -0.5 cm, and the change rate is (1.5 - 2) ÷ 2 × 100% = -25%; the change amount of the diameter of the lymph node is 0.8 - 1 = -0.2 cm, and the change rate is (0.8 - 1) ÷ 1 × 100% = -20%. In terms of the metabolic parameter feature change information, the change amount of the metabolic value of the primary tumor is 2.8 - 3.5 = -0.7, and the change rate is (2.8 - 3.5) ÷ 3.5 × 100% = -20%; the change amount of the metabolic value of the lymph node is 2.0 - 2.5 = -0.5, and the change rate is (2.0 - 2.5) ÷ 2.5 × 100% = -20%. These change information reflect the changes in the size and metabolic activity of the tumor and lymph nodes within the preset time period.

[0045] Process the clinical parameter information, lymph node residual degree information, imaging feature change information and metabolic parameter feature change information of the target user to generate the lymph node residual association information of the target user. Integrate the patient's clinical parameter information (age 58 years old, male, with a 30 - year smoking history, tumor clinical stage IIIB, PD - L1 expression level 50%, pathological type adenocarcinoma), lymph node residual degree information (moderate residual), imaging feature change information (change amount and change rate of the diameters of the primary tumor and lymph nodes), and metabolic parameter feature change information (change amount and change rate of the metabolic values of the primary tumor and lymph nodes). Use data analysis algorithms to comprehensively consider the associations between various factors. For example, it is found that for this patient with a long - term smoking history and a relatively late tumor clinical stage, although the continuous decrease in the diameter of the lymph node and the metabolic value has certain positive significance, considering the current situation of moderate residual, it still indicates that the risk of tumor recurrence cannot be ignored. Through these comprehensive analyses, generate the lymph node residual association information containing the associations of multiple factors.

[0046] Process the lymph node residue correlation information of the target user to generate the adjuvant treatment intervention information of the target user. Based on the above lymph node residue correlation information, if the degree of lymph node residue in the patient is moderate, and the size and metabolic activity of the tumor and lymph nodes have improved but there are still risks within the preset time period. The doctor can consider formulating an enhanced adjuvant chemotherapy plan for the patient. For example, on the basis of the original paclitaxel and carboplatin chemotherapy plan, appropriately increase the dose of paclitaxel by 20%, and extend the chemotherapy cycle from the original 4 cycles to 6 cycles to further eliminate the potentially remaining tumor cells. At the same time, considering that the patient's PD-L1 expression level is 50%, immunotherapy drugs such as pembrolizumab can be considered in combination to enhance the killing effect of the patient's own immune system on tumor cells, improve the treatment effect, and reduce the risk of tumor recurrence.

[0047] In another implementation, process the imaging parameter information and metabolic parameter information of the target user within the preset time period to generate the access value of the first frequency and the moment state vector of the second frequency. Within the preset time period (1 - 3 months), perform multiple imaging and metabolic parameter detections on the patient. Assume that at the first detection (representing the first frequency), data is obtained through chest CT and PET / CT scans. The diameter of the primary tumor after treatment is measured to be 2 cm (this can be used as a quantitative index of tumor size, corresponding to in the formula), and at this time, the metabolic value of the primary tumor (SUVmax) is 3.5 (corresponding to the quantitative index of tumor metabolic activity, which can be regarded as part of in the formula). Use these data as the access value of the first frequency.

[0048] For the moment state vector of the second frequency, we consider correlating the data of the second detection (assumed to be performed in the second month) with the data of the first detection for representation. For example, at the second detection, the diameter of the primary tumor becomes 1.8 cm and the metabolic value (SUVmax) becomes 3.2. To generate the moment state vector, calculate the change amount and some related statistical features (such as the change rate, etc.) between the two detections. The change amount of the primary tumor diameter is 1.8 - 2 = -0.2 cm, and the change rate is (1.8 - 2) ÷ 2 × 100% = -10%; the change amount of the metabolic value is 3.2 - 3.5 = -0.3, and the change rate is (3.2 - 3.5) ÷ 3.5 × 100% ≈ -8.6%. Integrate these change amounts, change rates, and other information to form the moment state vector of the second frequency.

[0049] Process the access value of the first frequency and the moment state vector of the second frequency to generate the moment state vector of the first frequency. The method includes the calculation formula for obtaining the moment state vector of the first frequency. The calculation formula is: ; Among them, Represents the moment state vector of the first frequency calculated at time point t, which represents the value of the th tumor size quantification index in the first frequency access value at time point t, which represents the value of the th tumor size quantification index in the first frequency access value at time point t, which represents the value of the th tumor metabolic activity quantification index in the moment state vector of the second frequency at time point t, which represents the value of the th tumor metabolic activity quantification index in the moment state vector of the second frequency at time point t, which represents the weight coefficient used to adjust , which represents the weight coefficient used to adjust , which represents the weight coefficient used to adjust , which represents the weight coefficient used to adjust , which represents the bias term of the first half of the formula, which represents the bias term of the second half of the formula.

[0050] Assume that the parameters obtained through training with a large number of samples are: .

[0051] Taking the tumor size quantification index (such as the diameter of the primary tumor) and the tumor metabolic activity quantification index (such as the metabolic value of the primary tumor) as examples, calculate the moment state vector of the first frequency. Here, assume that both n and m are 1, i = 1, and j = 1. is 2 cm, is 3.5, is 2, is the calculated value corresponding to the primary tumor metabolic value in the moment state vector of the second frequency (assume it is the change rate calculated above -8.6% converted to a decimal -0.086).

[0052] First, calculate the exponential part of the formula: .

[0053] Then The partial calculation is: .

[0054] .

[0055] . Therefore, the moment state vector of the first frequency It is: 。

[0056] Generate the image feature change information and metabolic parameter feature change information of the target user based on the moment state vector of the first frequency and the moment state vector of the second frequency. Combine the moment state vector of the first frequency ( ), and the moment state vector of the second frequency to generate the image feature change information and metabolic parameter feature change information. Suppose we set a rule that if the value of the moment state vector of the first frequency is greater than a certain threshold (such as 0.5, which is determined based on the analysis of a large number of samples), and both the tumor diameter and the metabolic value change rate in the moment state vector of the second frequency are negative numbers (indicating that the tumor is shrinking and the metabolic activity is decreasing), then it is considered that the changes in the image features and metabolic parameter features show a positive trend.

[0057] For this patient, and the change rate of the primary tumor diameter is -10%, and the change rate of the metabolic value is -8.6%, both of which are negative numbers. From this, it can be judged that within the preset time period, the image feature change information of this patient shows that the tumor size has a shrinking trend, and the metabolic parameter feature change information shows that the tumor metabolic activity has also decreased, which indicates that to a certain extent, it plays a role in inhibiting tumor growth.

[0058] The server first obtains the clinical, imaging, and metabolic parameter information of the target user, as well as the training sample set and the preset model. Process the preset model with the training sample set, and through operations such as data feature acquisition, sampling, grouping, and missing value imputation, generate a prediction model for the pathological remission rate of the target metastatic lymph node. Then, extract features from the multi-source parameter information and perform regression analysis to obtain risk prediction factors including the diameter of the primary tumor after treatment, etc. Input the risk prediction factors into the target model, use a specific formula to calculate the predicted value of the lymph node pathological remission status, and then obtain the lymph node residual information. Finally, perform quantitative analysis on the lymph node residual information, combine the image and metabolic parameter change information within the preset time period, and clinical parameters to generate lymph node residual association information, and accordingly formulate adjuvant treatment intervention information.

[0059] In one implementation manner, as Figure 2 shown, the present application also provides a prediction device for the pathological state of lymph nodes after immunotherapy, including:

[0060] An acquisition module 201, configured to acquire the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, the training sample set, and the preset prediction model for the pathological remission rate of metastatic lymph nodes, where the target user is a user with lymphatic metastatic lung cancer, and the training sample set is patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment;

[0061] The processing module 202 is configured to process the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; process the clinical parameter information, the imaging parameter information, and the metabolic parameter information of the target user to generate a risk prediction factor; process the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; and process the lymph node residue information of the target user to generate the adjuvant treatment intervention information of the target user.

[0062] In another implementation manner of the present application, the processing module 202 is configured to process the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes, including:

[0063] Obtain the data features of any quantity in the training sample set;

[0064] Generate a sampling ratio based on the quantities of the data features in the training sample set;

[0065] Sample the training sample set based on the sampling ratio to generate a preset quantity of sampling features;

[0066] Process any data feature and each sampling feature to generate a response group and a poor response group, where each data group contains a preset quantity of data samples, and at least one data sample includes identification information;

[0067] Perform missing value imputation processing on the response group and the poor response group respectively to generate missing value imputation prediction information;

[0068] Train the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the data samples in the response group and the poor response group to generate a trained prediction model for the pathological remission rate of metastatic lymph nodes;

[0069] Process the trained prediction model for the pathological remission rate of metastatic lymph nodes to generate a verification result;

[0070] If the data sample containing identification information in the verification result represents a risk factor affecting lymphatic metastatic lung cancer, then use the trained prediction model for the pathological remission rate of metastatic lymph nodes as the target prediction model for the pathological remission rate of metastatic lymph nodes.

[0071] In another implementation manner of the present application, the processing module 202 is configured to process the initial prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set and the verification sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes, and further includes:

[0072] Obtain the missing sample information of the response group and the poor response group respectively;

[0073] Process the response group and the poor response group respectively based on the missing sample information of the response group and the poor response group, and generate predicted values for the missing samples;

[0074] Obtain the true value corresponding to the missing sample;

[0075] Process the predicted value of the missing sample and the true value corresponding to the missing sample to generate missing value imputation prediction information;

[0076] The method further includes a calculation formula for obtaining the missing value imputation prediction information, and the calculation formula is:

[0077] 。

[0078] Wherein, represents the constant term of the formula, represents the regression coefficient related to clinical parameters, represents the value of the k-th clinical parameter, represents the regression coefficient related to imaging parameters, represents the value of the m-th imaging parameter, represents the regression coefficient related to metabolic parameters, represents the value of the n-th metabolic parameter.

[0079] In another implementation manner of the present application, the processing module 202 is configured to process the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor, including:

[0080] Perform feature extraction processing on the clinical parameter information of the target user to generate clinical parameter features of the target user, wherein the clinical parameter features of the target user include age, gender, smoking history, tumor clinical stage, PD-L1 expression level, and pathological type;

[0081] Perform feature extraction processing on the imaging parameter information of the target user to generate imaging features of the target user, wherein the imaging features of the target user include the diameter of the primary tumor before treatment, the diameter of the primary tumor after treatment, the change rate of the diameter of the primary tumor before and after treatment, the diameter of the lymph node before treatment, the diameter of the lymph node after treatment, and the change rate of the diameter of the lymph node before and after treatment;

[0082] Perform feature extraction processing on the metabolic parameter information of the target user to generate the metabolic parameter features of the target user. Among them, the metabolic parameter features of the target user include the pre-treatment primary tumor metabolic value, the post-treatment primary tumor metabolic value, the change rate of the primary tumor metabolic value before and after treatment, the pre-treatment lymph node metabolic value, the post-treatment lymph node metabolic value, and the change rate of the lymph node metabolic value before and after treatment;

[0083] Perform regression analysis processing on the clinical parameter features of the target user, the imaging features of the target user, and the metabolic parameter features of the target user to generate risk prediction factors. Among them, the risk prediction factors include the post-treatment primary tumor diameter, the reduction rate of the primary tumor diameter, the change amount of the maximum standardized uptake value of the primary tumor, and the change amount of the maximum standardized uptake value of the lymph node.

[0084] In another implementation manner of the present application, the processing module 202 is configured to process the risk prediction factors based on the target metastatic lymph node pathologic remission rate prediction model to generate the lymph node residual information of the target user, including:

[0085] Process the risk prediction factors based on the target metastatic lymph node pathologic remission rate prediction model to generate a predicted value of the lymph node pathologic remission status;

[0086] Process the predicted value of the lymph node pathologic remission status to generate the lymph node residual information of the target user;

[0087] The method further includes a calculation formula for obtaining the predicted value of the lymph node pathologic remission status, and the calculation formula is:

[0088] .

[0089] In another implementation manner of the present application, the processing module 202 is configured to process the lymph node residual information of the target user to generate the adjuvant treatment intervention information of the target user, including:

[0090] Perform quantitative analysis processing on the lymph node residual information of the target user to generate lymph node residual degree information;

[0091] Obtain the imaging parameter information and metabolic parameter information of the target user within a preset time period;

[0092] Process the imaging parameter information and metabolic parameter information of the target user within the preset time period to generate imaging feature change information and metabolic parameter feature change information;

[0093] Process the clinical parameter information, lymph node residue degree information, imaging feature change information, and metabolic parameter feature change information of the target user to generate the lymph node residue association information of the target user;

[0094] Process the lymph node residue association information of the target user to generate the adjuvant treatment intervention information of the target user.

[0095] In another implementation manner of the present application, the processing module 202 is configured to process the imaging parameter information and metabolic parameter information of the target user within a preset time period to generate imaging feature change information and metabolic parameter feature change information, including:

[0096] Process the imaging parameter information and metabolic parameter information of the target user within a preset time period to generate the first-frequency access value and the second-frequency moment state vector;

[0097] Process the first-frequency access value and the second-frequency moment state vector to generate the first-frequency moment state vector;

[0098] Generate the imaging feature change information and metabolic parameter feature change information of the target user based on the first-frequency moment state vector and the second-frequency moment state vector;

[0099] The method includes a calculation formula for obtaining the first-frequency moment state vector, and the calculation formula is:

[0100] ;

[0101] Wherein, represents the first-frequency moment state vector calculated at time point t, represents the value of the th tumor size quantization index in the first-frequency access value at time point t, represents the value of the th tumor size quantization index in the first-frequency access value at time point t, represents the value of the th tumor metabolic activity quantization index in the second-frequency moment state vector at time point t, represents the value of the th tumor metabolic activity quantization index in the second-frequency moment state vector at time point t, represents the weight coefficient used to adjust , represents the weight coefficient used to adjust , represents the weight coefficient used to adjust , represents the weight coefficient used for adjustment and represents the bias term of the first half of the formula, and

[0102] Each embodiment in this application is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the embodiments of the prediction method, electronic device, electronic equipment, and readable storage medium for evaluating the pathological state of lymph nodes after immunotherapy, since they are basically similar to the embodiments of the prediction method for the pathological state of lymph nodes after immunotherapy described above, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the embodiments of the prediction method for the pathological state of lymph nodes after immunotherapy described above.

Claims

1. A method for predicting the pathological status of lymph nodes after immunotherapy, characterized in that, Including: Obtaining the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, a training sample set, and a preset prediction model for the pathological remission rate of metastatic lymph nodes, where the target user is a user with lymphatic metastatic lung cancer, and the training sample set is patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment; Processing the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; Processing the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor; Processing the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; Processing the lymph node residue information of the target user to generate the adjuvant treatment intervention information of the target user, including: performing quantitative analysis processing on the lymph node residue information of the target user to generate the lymph node residue degree information; obtaining the imaging parameter information and metabolic parameter information of the target user within a preset time period; processing the imaging parameter information and metabolic parameter information of the target user within the preset time period to generate the imaging feature change information and the metabolic parameter feature change information; processing the clinical parameter information, the lymph node residue degree information, the imaging feature change information, and the metabolic parameter feature change information of the target user to generate the lymph node residue association information of the target user; processing the lymph node residue association information of the target user to generate the adjuvant treatment intervention information of the target user.

2. The method according to claim 1, wherein Processing the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes, including: Obtaining any quantity of data features in the training sample set; Generating a sampling ratio based on the quantity of each data feature in the training sample set; Sampling the training sample set based on the sampling ratio to generate a preset quantity of sampling features; Processing any data feature with each sampling feature to generate a response group and a non - response group, where each data group contains a preset quantity of data samples, and at least one data sample includes identification information; Performing missing value imputation processing on the response group and the non - response group respectively to generate missing value imputation prediction information; Training the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the data samples in the response group and the non - response group to generate a trained prediction model for the pathological remission rate of metastatic lymph nodes; Processing the trained prediction model for the pathological remission rate of metastatic lymph nodes to generate a verification result; If the data sample containing identification information in the verification result represents a risk factor affecting lymphatic metastatic lung cancer, then using the trained prediction model for the pathological remission rate of metastatic lymph nodes as the target prediction model for the pathological remission rate of metastatic lymph nodes.

3. The method according to claim 2, wherein Processing the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes, further including: Obtain the missing sample information of the response group and the poor response group respectively; Process the response group and the poor response group respectively based on the missing sample information of the response group and the poor response group to generate predicted values for the missing samples; Obtain the true values corresponding to the missing samples; Process the predicted values of the missing samples and the true values corresponding to the missing samples to generate missing value imputation prediction information; The method further includes a calculation formula for obtaining the missing value imputation prediction information, and the calculation formula is: ； Among them, represents the constant term of the formula, represents the regression coefficient related to clinical parameters, represents the value of the k-th clinical parameter, represents the regression coefficient related to imaging parameters, represents the -th value of the imaging parameter, represents the regression coefficient related to metabolic parameters, represents the -th value of the metabolic parameter.

4. The method according to claim 1, wherein Process the clinical parameter information, the imaging parameter information, and the metabolic parameter information of the target user to generate risk prediction factors, including: Perform feature extraction processing on the clinical parameter information of the target user to generate the clinical parameter features of the target user. Among them, the clinical parameter features of the target user include age, gender, smoking history, tumor clinical stage, PD-L1 expression level, and pathological type; Perform feature extraction processing on the imaging parameter information of the target user to generate the imaging features of the target user. Among them, the imaging features of the target user include the diameter of the primary tumor before treatment, the diameter of the primary tumor after treatment, the change rate of the diameter of the primary tumor before and after treatment, the diameter of the lymph node before treatment, the diameter of the lymph node after treatment, and the change rate of the diameter of the lymph node before and after treatment; Perform feature extraction processing on the metabolic parameter information of the target user to generate the metabolic parameter features of the target user. Among them, the metabolic parameter features of the target user include the metabolic value of the primary tumor before treatment, the metabolic value of the primary tumor after treatment, the change rate of the metabolic value of the primary tumor before and after treatment, the metabolic value of the lymph node before treatment, the metabolic value of the lymph node after treatment, and the change rate of the metabolic value of the lymph node before and after treatment; Perform regression analysis processing on the clinical parameter features, the imaging features, and the metabolic parameter features of the target user to generate risk prediction factors. Among them, the risk prediction factors include the diameter of the primary tumor after treatment, the reduction rate of the diameter of the primary tumor, the change amount of the maximum standardized uptake value of the primary tumor, and the change amount of the maximum standardized uptake value of the lymph node; 5. The method according to claim 4, wherein Process the risk prediction factors based on the target metastatic lymph node pathological remission rate prediction model to generate the lymph node residue information of the target user, including: Process the risk prediction factors based on the target metastatic lymph node pathological remission rate prediction model to generate a predicted value of the pathological remission state of the lymph node; Process the predicted value of the pathological remission state of the lymph node to generate the lymph node residue information of the target user; The method further includes a calculation formula for obtaining the predicted value of the pathological remission state of the lymph node, and the calculation formula is: 。 6. The method according to claim 1, wherein Process the imaging parameter information and the metabolic parameter information of the target user within a preset time period to generate imaging feature change information and metabolic parameter feature change information, including: Process the imaging parameter information and the metabolic parameter information of the target user within a preset time period to generate the access value of the first frequency and the moment state vector of the second frequency; Process the access value of the first frequency and the moment state vector of the second frequency to generate the moment state vector of the first frequency; Generate the image feature change information and metabolic parameter feature change information of the target user based on the moment state vector of the first frequency and the moment state vector of the second frequency; The method includes a calculation formula for obtaining the moment state vector of the first frequency, and the calculation formula is: ； Among them, represents the moment state vector of the first frequency calculated at time point t, represents the value of the -th tumor size quantization index in the first frequency access value at time point t, represents the value of the -th tumor size quantization index in the first frequency access value at time point t, represents the value of the -th tumor metabolic activity quantization index in the moment state vector of the second frequency at time point t, represents the value of the -th tumor metabolic activity quantization index in the moment state vector of the second frequency at time point t, represents the weight coefficient used to adjust , represents the weight coefficient used to adjust , represents the weight coefficient used to adjust , represents the weight coefficient used to adjust , represents the bias term of the first half of the formula, represents the bias term of the second half of the formula.

7. A prediction device for the pathological state of lymph nodes after immunotherapy, characterized in that, For implementing the method described in claim 1, the apparatus includes: An acquisition module, configured to acquire the clinical parameter information of the target user, the imaging parameter information of the target user, the metabolic parameter information of the target user, a training sample set, and a preset prediction model for the pathological remission rate of metastatic lymph nodes, wherein the target user is a user with lymphatic metastatic lung cancer, and the training sample set is patients with lymphatic metastatic lung cancer who have completed lymph node metastasis treatment; A processing module, configured to process the preset prediction model for the pathological remission rate of metastatic lymph nodes based on the training sample set to generate a target prediction model for the pathological remission rate of metastatic lymph nodes; process the clinical parameter information of the target user, the imaging parameter information of the target user, and the metabolic parameter information of the target user to generate a risk prediction factor; process the risk prediction factor based on the target prediction model for the pathological remission rate of metastatic lymph nodes to generate the lymph node residue information of the target user; process the lymph node residue information of the target user to generate the adjuvant treatment intervention information of the target user.

8. An electronic device, characterized in that, Comprising: A first processor; And a memory for storing the executable instructions of the first processor; Wherein, the first processor is configured to execute the prediction method for the pathological state of lymph nodes after immunotherapy described in any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a second processor, it implements the prediction method for the pathological state of lymph nodes after immunotherapy described in any one of claims 1 to 6.

Images