Gastric cancer immunotherapy pre-evaluation method based on artificial intelligence method

Abstract

The invention discloses a gastric cancer immunotherapy pre-evaluation method based on an artificial intelligence method, and relates to the field of calculation and prediction of gastric cancer immunotherapy curative effect, and the method comprises the following steps: S100, analyzing data of a gastric cancer patient receiving immunotherapy, and constructing an immune feature data set; s200, compiling a Python program, and screening the features based on a PPI-RWR algorithm and a random gene perturbation algorithm; s300, amplifying the sample by using a data enhancement algorithm based on a diffusion model; s400, performing modeling prediction on the standardized data obtained in the step S300 by using six machine learning algorithms, and finding optimal parameters of each model through ten-fold cross validation and grid parameter searching; s500, each machine learning algorithm executes ten-fold cross validation under the optimal hyper-parameter, an optimal machine learning model is selected through comparison, and a final model is constructed; and S600, constructing an external test set to verify the model obtained in the step S400.

Description

Technical Field

[0001] The present invention relates to the field of computational prediction of gastric cancer immunotherapy efficacy, and in particular to a gastric cancer immunotherapy pre-evaluation method based on artificial intelligence methods. Background Art

[0002] In recent years, with the progress of immunotherapy research and the exploration of new immunotherapy targets in gastric cancer, major breakthroughs have been made in the treatment of gastric cancer. Compared with current traditional treatments, immunotherapy has significantly improved the survival benefits of gastric cancer patients, breaking the previous monopoly of chemotherapy and targeted therapy. However, the biggest challenge of this therapy is that the efficacy of patients varies greatly, and only a part of patients can benefit from it. Therefore, it is an urgent problem to accurately predict the patient groups who can benefit from immunotherapy, formulate corresponding individualized treatment plans, and avoid the harm and burden of improper treatment on patients.

[0003] A major challenge in predicting the efficacy of immunotherapy in patients is to find reliable predictive markers from patients who have received immunotherapy. The current mainstream method is to perform auxiliary diagnostic tests by experimentally detecting known biomarkers (tumor mutation load, PD-1 / PD-L1, peripheral blood biomarkers, etc.). However, traditional experiments are time-consuming and labor-intensive, and the detection efficiency can no longer meet the needs of related research. With the in-depth research in related fields, the existing biomarkers can no longer meet the growing demand. Studies have reported that there is no obvious correlation between PD-L1 expression and ICI treatment response. Some studies have even found that patients who respond to immunotherapy have lower PD-L1 expression levels. Litchfield et al. recently found that traditional biomarkers can only predict about 60% of immunotherapy responses, indicating that new biomarkers are still to be discovered. Therefore, it is necessary to develop effective methods to identify biomarkers from patients who have received immunotherapy, and ultimately maximize the benefits of immunotherapy. With the rapid development of computer technology, machine learning has begun to be applied to biomarker screening with its flexibility and strong learning ability. CHEN et al. used a support vector machine model to establish a lncRNAs model for automatic classification of microsatellite instability consisting of 16 lncRNA features, which may become targets for PD-1 / PD-L1 immunotherapy. WANG et al. used a random forest algorithm to identify a microbial feature in 32 bacterial genera associated with early gastric cancer. SAITO et al. used LC / ESI-MS to determine the levels of 236 lipid molecules such as phosphatidylcholine and sphingomyelin in 1592 plasma samples and used a logistic regression algorithm for discrimination. The results showed that the accuracy of gastric cancer diagnosis was as high as 94.4%. However, while these previous methods have been successful, they often ignore the correlation and interaction between genes. Genes do not work independently, and the complex network interactions between them have an important impact on the occurrence, development and response to treatment of the disease. Ignoring this interaction may result in the screened features being limited to a single gene level and unable to fully reflect the potential synergistic effect. Therefore, it is particularly important to use modern complex network theory and combine the interactions between genes to conduct a more comprehensive analysis of the efficacy of immunotherapy.

[0004] Another major challenge in predicting the efficacy of immunotherapy in patients is that immunotherapy experiments are very expensive and have a long experimental cycle, resulting in a small amount of available valid data samples, which greatly affects the accuracy of the prediction model. Directly applying machine learning is bound to cause problems such as poor model prediction performance and serious overfitting. It is necessary to develop corresponding technologies to break the limitation of data volume. At present, data augmentation technology has achieved great success in the field of computer vision. It reduces the risk of model overfitting and improves the generalization performance of the model. It has become a common method in the field of computer vision. However, biomedical data is very different from image data and cannot be used to implement data augmentation operations such as image flipping. Therefore, in the analysis of high-throughput sequencing data in the biomedical field, especially in gene expression data, there are few reports on data generation and data augmentation.

[0005] In summary, given the scarcity of immunotherapy data samples, the complex interactions between genes, and the difficulty of existing biomarkers in fully predicting treatment effects, it is particularly necessary to develop a method that can quickly and accurately predict the efficacy of immunotherapy. This can not only overcome the shortcomings of traditional methods in marker screening and data volume limitations, but also provide a theoretical basis for individualized immunotherapy programs, thereby further guiding clinical decision-making and efficacy evaluation, and maximizing the benefits of immunotherapy. Summary of the invention

[0006] The purpose of the present invention is to provide a pre-evaluation method for gastric cancer immunotherapy based on an artificial intelligence method, and propose a feature screening strategy of a protein interaction restart random walk (Protein-ProteinInteraction Random Walk With Restart, PPI-RWR) algorithm combined with a random network walk and a random gene perturbation (Random Gene Perturbation, RGP) algorithm, and introduce a data enhancement method combining a generative diffusion model and active learning, which is applied to the pre-evaluation of the efficacy of gastric cancer immunotherapy. Through this method, while effectively solving the problems of insufficient data and limited sample size in traditional experimental determinations, biomarkers closely related to the efficacy of immunotherapy can be screened out, and the excellent extrapolation performance of the model can be ensured. This method not only provides new ideas for the prediction of the efficacy of immunotherapy, but also provides reliable support for clinical treatment decisions, and helps the development of individualized and precise immunotherapy.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A new method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence methods, including:

[0009] Step S100, analyze the data of gastric cancer patients receiving immunotherapy and construct an immune signature data set. RNA-seq data of 78 gastric cancer patients who received immunotherapy were downloaded from ICBaltas. The treatment effects of these patients were determined by the RECIST criteria. In order to ensure the effect of subsequent modeling, the data was preprocessed: data filtering, missing value filling, data standardization, and data conversion. Subsequently, the differentially expressed genes between patients with and without treatment response were extracted, and finally the data were summarized as behavioral genes and listed as a matrix of samples.

[0010] Step S200, write a Python program to screen features based on the PPI-RWR algorithm and the random gene perturbation algorithm. Select known target genes and pathway genes related to immunotherapy as seed genes for network propagation, and select genes with the top 10% propagation scores. Finally, continue to screen biomarkers through the random gene perturbation algorithm to further narrow the feature space and obtain the optimal gene set for subsequent model construction.

[0011] Step S300: Amplify the sample using a data enhancement algorithm based on a diffusion model. This technology applies the diffusion model to immunotherapy sample generation for the first time, and innovatively combines active learning with data enhancement, taking into account the quality of new data during data amplification, effectively solving the limitation of the scarce amount of immunotherapy data;

[0012] Step S400, use six machine learning algorithms to model and predict the standardized data obtained in step S300, including: K-nearest Neighbors (KNN), support vector machine (SVM), decision tree (DT), random forest (RF), extreme gradient boosting (XGB), and logistic regression (LR). The optimal parameters of each model are found through ten-fold cross validation and grid parameter search.

[0013] Step S500: Perform ten-fold cross validation on each machine learning algorithm under the optimal hyperparameters, compare and select the optimal machine learning model, and build the final model.

[0014] S600: Build an external test set to verify the model obtained in step S400 to prove the good generalization ability of the model.

[0015] The step S100 specifically includes:

[0016] Step S110: Download RNA-seq data of gastric cancer patients who have received immunotherapy from ICBaltas as an initial data set.

[0017] Step S120, preprocess the data: use Python Pandas package to implement data filtering, missing value filling, and min-max data standardization.

[0018] Step S130: Using the R package limma, the differentially expressed genes were screened and retained according to the conditions of |log2FC|≥1 and FDR<0.05.

[0019] Step S140: In order to compress the data value range, smooth the distribution and improve the stability of the model, the data is logarithmized using the Python Numpy package.

[0020] The step S200 specifically includes:

[0021] Step S210: construct a target network. Download the human PPI network from the STRING database v.12.0.

[0022] Step S220, select existing immunotherapy target genes (PD-1, PD-L1, CTLA-4) and immune-related pathway (T cell receptor, B cell receptor, antigen presentation) genes as seed genes, use PythonNetworkX's page-rank method, and determine immunotherapy-related genes through the PPI-RWR algorithm.

[0023] Step S230, iteratively running the random gene perturbation algorithm until the evaluation index begins to decrease, and continuing to screen key biomarkers.

[0024] The step S300 specifically includes:

[0025] Step S310: Use the data set D0 obtained in step S200 to construct an agent model C0 for predicting the efficacy of immunotherapy.

[0026] Step S320: Run the diffusion model to perform data enhancement to obtain 500 enhanced samples, which constitute the sample pool U0 required for active learning.

[0027] Step S330: Use the proxy model obtained in step S310 to predict the sample pool obtained in step S320 and calculate the expected improvement value EI of each sample to be selected, sort the expected improvement value of each sample to be selected in descending order, select the top 50 samples and add them to the original data set D0 to form the enhanced data set D1. At this time, there are 450 samples left in the sample pool U1 to be selected;

[0028] Step S340: Retrain the enhanced dataset D1 in step S330 to obtain an updated proxy model C1. Use the proxy models C1 and C1 to predict the sample pool U1 and calculate the expected improvement EI. After sorting in descending order, select the top 50 samples and add them to the dataset D1 to form the enhanced dataset D2. At this time, there are 400 samples left in the sample pool U2 to be selected.

[0029] Step S350: Retrain the enhanced dataset D2 in step S340 to obtain an updated proxy model C2, and make predictions on the sample pool U2 and calculate the expected improvement EI. After sorting in descending order, select the top 50 samples and add them to the dataset D2 to form the enhanced dataset D3. At this time, there are 350 samples left in the sample pool U3 to be selected.

[0030] Step S360: Use the enhanced data set D3 in step S330 to retrain to obtain the updated proxy model C3 and predict the sample pool U3 and calculate the expected improvement value EI. After sorting in descending order, select the top 50 samples and add them to the data set D3 to form the enhanced data set D4. At this time, there are 300 samples left in the sample pool U4 to be selected.

[0031] The step S400 specifically includes:

[0032] Step S410: determining the parameter optimization space of each machine learning algorithm;

[0033] Step S420, divide the data set according to ten-fold cross validation to ensure that the distribution of each data is as uniform as possible, where 9 data are used as training sets and the remaining 1 data is used as a test set;

[0034] Step S430, when executing the parameter optimization process, taking XGB as an example, for each set of hyperparameters, first train the model on the training set of step S420, and calculate the evaluation index of the classification model on the test set. The whole process is looped 10 times to ensure that each piece of data is used as a training set and a test set, and finally 10 sets of evaluation index results are obtained. The average value of these 10 sets of results is taken as the performance of the model under this set of hyperparameters. Next, traverse all the hyperparameter combinations defined in step S410, calculate the evaluation index of each combination, and finally select the set of hyperparameters with the best performance as the optimal hyperparameters of XGB. Other machine learning algorithms follow the same steps to determine their respective optimal hyperparameters.

[0035] Step S440: Execute steps S420 and S430 on the enhanced data sets D2, D3, and D4 in sequence to obtain the optimal hyperparameters of each machine learning algorithm under different data sets.

[0036] The step S500 specifically includes:

[0037] Step S510: Based on the enhanced data set D1, ten-fold cross validation is performed in turn for each machine learning algorithm according to the optimal hyperparameters obtained in step S400 to obtain ten average evaluation indicators R2, RMSE and MAE, and the optimal model is selected according to the criteria of maximum R2 and minimum RMSE and MAE;

[0038] Step S520: Similarly, step S510 is performed on the enhanced data sets D2, D3, and D4 in sequence to select the optimal model corresponding to each data set;

[0039] Step S530, comparing the model performances under different enhanced data sets obtained in step S510 and step S520, and selecting the optimal enhanced data set and the optimal machine learning algorithm;

[0040] Step S540: According to the comparison result of step S530, it is determined that the optimal enhanced data set is D4 and the optimal friction sensitivity prediction algorithm is XGB. Therefore, the final immunotherapy efficacy prediction model is constructed based on all the data of D4 using the optimal hyperparameters.

[0041] The step S600 specifically includes:

[0042] S610. Collect gastric cancer patient samples from the GEO database. First, use the combat package of R language to eliminate the batch effect. Then use the TIDER package of R language to simulate the results of patients after immunotherapy, and construct an external test set for subsequent verification.

[0043] S620 , using the optimal XGB model constructed in step S540 to predict the external test set generated in step S610 , and calculating the evaluation index of the model to evaluate the generalization ability of the model.

[0044] The beneficial effects of a gastric cancer immunotherapy pre-evaluation method based on an artificial intelligence method disclosed in this application include but are not limited to:

[0045] 1. The present invention innovatively introduces the PPI-RWR algorithm and the RGP algorithm into the feature screening for predicting the efficacy of immunotherapy. Starting from the target gene, PPI-RWR explores the interaction between genes through random walks, reveals potential synergistic effects, and identifies genes closely related to immunotherapy from the global network. This method not only considers the importance of individual genes, but also captures the complex interactions between genes, significantly improving the biological significance of the screening results.

[0046] 2. Aiming at the limitation of insufficient sample size of immunotherapy data, this paper proposes AL-Diffusion data enhancement technology for the first time, combining generative deep diffusion model with active learning. The diffusion model generates new samples that are close to real data, and active learning monitors their quality to ensure the effectiveness of data enhancement and avoid the excessive reliance of conventional active learning on experimental annotation.

[0047] 3. The invention has a wide range of applications, is easy to operate, and can make fast and accurate predictions. The optimal model constructed based on 79 gene features is not only suitable for immunotherapy research, but can also be widely used in other related fields, and can quickly and accurately predict through gene expression values. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of the present invention,

[0049] Figure 2 This is an example diagram of the results of the data processing part of the present invention.

[0050] Figure 3 This is a schematic diagram of the core algorithm PPI-RWR algorithm of the present invention.

[0051] Figure 4 This is an example diagram of the PPI-RWR algorithm results.

[0052] Figure 5 This is an example diagram of the RGP algorithm results.

[0053] Figure 6 This is a schematic diagram of the AL-Diffusion data enhancement technology, the core technology of the present invention.

[0054] Figure 7 Schematic diagram of the diffusion model used in the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clear, the present application is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0056] On the contrary, the present application covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present application as defined by the claims. Further, in order to make the public have a better understanding of the present application, some specific details are described in detail in the detailed description of the present application below. Those skilled in the art can fully understand the present application without the description of these details.

[0057] The following is a detailed description of a method for monitoring the growth process and quality of peach blossoms involved in the embodiments of the present application. It should be noted that the following embodiments are only used to explain the present application and do not constitute a limitation of the present application.

[0058] like Figure 1 As shown, a pre-evaluation method for gastric cancer immunotherapy based on artificial intelligence method comprises:

[0059] Step 1: Download gene expression data of gastric cancer patients who have received immunotherapy from ICBaltas. The data is pre-processed by pre-data filtering, missing value filling, data standardization, and data conversion. Subsequently, differentially expressed genes between patients with and without treatment response are extracted. Data with a large numerical scale range will not only greatly slow down the training process, but also seriously affect the model performance and cause biased results. Therefore, the data is min-max standardized according to eq(1) to obtain the data set, and the data is logarithmized:

[0060]

[0061] The final dataset is as follows: Figure 2 As shown, the rows of the dataset represent genes and the columns represent patient samples.

[0062] Step 2: Use PPI-RWR technology and RGP algorithm to screen important immune-related genes according to eq (2), such as Figure 3 Network propagation is performed as shown:

[0063] P t+1 =(1-r)WP t +rP0(2)

[0064] Where: P t represents the probability vector of the node at step t, P0 represents the initial probability vector, r is the restart probability, and W is the transfer matrix, i.e., the column-normalized adjacency matrix of the network. The walk process starts randomly from a set of nodes in the network, repeatedly spreading from the current node to randomly selected adjacent nodes, or returning to the starting node. Let the network propagation reach a stable state (P t and P t+1 The change between -30 ), the probability vector can represent the association scores of all genes in the network with the starting point. The results of the PPI-RWR algorithm are shown in the following example. Figure 4 .

[0065] Next, we first selected the top 10% genes in the association score as our candidate genes. Then we executed the random gene perturbation algorithm to screen biomarkers according to the following pseudo code to obtain the optimal gene set for subsequent modeling. Finally, after 10 rounds of iterations, 79 genes were selected as the optimal feature set. The results of the screening process are shown in the figure below. Figure 5 shown.

[0066] Algorithm: Random Gene Perturbation (RGP) algorithm

[0067] Input: gene expression dataset R, estimator E

[0068] Output: Screened biomarker gene set D

[0069] 1: Begin

[0070] 2: Input R into E and calculate the AUC value A of A;

[0071] 3: Randomly shuffle the expression value of gene g and calculate the new AUC value A';

[0072] 4: Calculate the importance score of g I = A'-A;

[0073] 5: Select genes G whose importance scores rank in the top 20%;

[0074] 6: Repeat

[0075] 7: Use feature recursive elimination to screen genes in G;

[0076] 8: Remove the last 10% of genes to obtain a new gene set G';

[0077] 9: UntilAUC value starts to decrease;

[0078] 10: Return the optimal biomarker gene set D;

[0079] 11: End

[0080] Step 3: Use AL-Diffusion data enhancement technology to amplify the data set to generate high-quality samples. Specifically, first perform the diffusion process according to eq (3) and add noise to disturb the original sample: See the schematic diagram of the core technology of the present invention, AL-Diffusion data enhancement technology. Figure 6 .

[0081]

[0082] Among them, x0 represents the original sample, x t represents the sample with noise added after the tth step, α tis a predefined noise scheduling parameter, ∈ is the Gaussian noise of the standard normal distribution, and through forward diffusion, the sample is gradually perturbed by the noise.

[0083] Then, the reverse diffusion process is performed according to eq (4) to finally generate new noise-free samples.

[0084]

[0085] in, It is the noise predicted by the neural network at step t, which helps to gradually remove noise, return the sample to a state close to the real distribution, and generate new high-quality new sample points. Through the reverse process, we can effectively restore the original features of the sample and generate enhanced samples that are highly similar to the original data.

[0086] Then, valuable samples are selected from the augmented samples through active learning technology, and the initial proxy model Cf is constructed using the original data set D0. The sample labels of the sample pool U0 are predicted and the expected improvement value EI of each sample to be selected is calculated according to eq(5)(6). The EI value of each sample to be selected is sorted in descending order, and the top 50 samples are selected and added to the original data set D0 to form the enhanced data set D1. At this time, there are 450 samples left in the sample pool U1 to be selected.

[0087] Then, valuable samples are selected from the augmented samples through active learning technology, and the initial proxy model Cf is constructed using the original data set D0. The sample labels of the sample pool U0 are predicted and the expected improvement value EI of each sample to be selected is calculated according to eq(5)(6). The EI value of each sample to be selected is sorted in descending order, and the top 50 samples are selected and added to the original data set D0 to form the enhanced data set D1. At this time, there are 450 samples left in the sample pool U1 to be selected.

[0088]

[0089]

[0090] Where σ is the standard deviation, is the probability density function, Φ(z) is the cumulative distribution function, μ * It refers to the maximum value expressed in the data set.

[0091] Step 4: Perform parameter optimization of different machine learning algorithms on the enhanced data set. Specifically: determine the parameter optimization space of each machine learning algorithm, as shown in Table 1:

[0092] Table 1. Hyperparameter optimization space for different machine learning algorithms

[0093]

[0094]

[0095] Combined with ten-fold cross validation, grid search was performed, and Acc, Pre, Recall, F1, and AUC were selected as model evaluation indicators. The indicator values ​​under each set of hyperparameters were calculated according to the following formula, and a set of hyperparameters corresponding to the optimal value was selected as the optimal hyperparameters:

[0096]

[0097]

[0098]

[0099]

[0100] Step 5: Compare the performance of various machine learning algorithms and select the best algorithm to build the final model. Specifically:

[0101] Comparing the impact of different augmentation times on model construction, as the number of data augmentation increases, the overall performance of the model shows a trend of gradual improvement. When 50 data augmentations are performed, the AUC value of most models is about 0.7. As the number of augmentations increases to 100 and 150 times, the performance of each model is improved to varying degrees, and the AUC value can reach more than 0.9. However, when the number of augmentations reaches 200 times, the performance improvement begins to flatten out. This may be because when the original sample size is small, too many augmentation operations will cause data redundancy in the sample space, thereby deviating from the actual distribution of the original data. Therefore, 200 data augmentations may be the best number of augmentations.

[0102] After determining the optimal number of enhancements, the performance of each machine learning algorithm was further compared. The Acc, Pre, Recall, F1, and AUC values ​​of each machine learning algorithm under 200 data enhancements were obtained by ten-fold cross validation. It can be seen from the results that XGB is the optimal immunotherapy efficacy prediction model. The hyperparameters and optimal results of the optimal model are shown in Table 2.

[0103] Table 2. Optimal model parameters and performance

[0104]

[0105] Step 6: According to step S600, an external test set is constructed to verify the optimal XGBoost model, and the analysis results are listed in Table 3. It can be seen that the constructed prediction model still shows good prediction performance in the external test set, which proves the effectiveness of our model.

[0106] Table 3. Results of external testing of the optimal model

[0107] Performance Indicators Acc Pre Recall F1 AUC Numeric 0.922 0.948 0.948 0.948 0.901

[0108] Step 7: Use the pickle.dump method to save the normalizer as a .dat file and the final model as a .pkl file, which can be loaded using the pickle.load method when predicting new samples.

[0109] In summary, the present invention achieves:

[0110] (i) Introduce PPI-RWR algorithm and RGP algorithm for feature screening.

[0111] A major challenge in predicting the efficacy of immunotherapy in patients is to find reliable predictive markers from patients who have received immunotherapy. The current mainstream method is to perform auxiliary diagnostic tests by detecting the expression of known biomarkers through immunohistochemistry experiments. However, the current detection efficiency can no longer meet the needs of related research, and with the in-depth research in related fields, the existing biomarkers can no longer meet the growing demand. It is urgent to discover new biomarkers to maximize the benefits of immunotherapy. Past studies often ignore the interactions between genes. Biological networks are a powerful resource for discovering genes or modules that drive the development of diseases. Modern complex network theory provides comprehensive tools to help us understand the occurrence and development of complex diseases such as tumors. At present, many bioinformatics technologies based on random walks, information diffusion and resistance have been proposed. They use networks for genetic analysis and have been successfully applied to identify pathogenic genes, disease treatment modules and drug targets. At present, such technologies have not been widely used in immunotherapy research.

[0112] In order to effectively identify the association between genes and find effective immune gene features, the present invention introduces the PPI-RWR algorithm into the immunotherapy research system. In a given PPI network, starting from the existing target genes, the network is walked with the same probability, and the genes are ranked according to the stabilized scores. Compared with previous methods, the algorithm we use takes into account the interactions between genes and reveals potential synergistic effects through the interactions between genes. This method has extremely high flexibility and scalability, and can quickly adapt to the growing needs of immunotherapy research.

[0113] In addition, in order to further extract key gene features, we proposed the RGP algorithm based on the PPI-RWR algorithm, which shuffles the gene expression values ​​and compares them with the original expression data to calculate the gene importance score, thereby determining the degree of influence of the gene in immunotherapy. Through this method, we can effectively identify genes that have a potential impact on the efficacy of immunotherapy and obtain a smaller and more accurate feature set, thereby greatly improving the subsequent calculation efficiency and accuracy.

[0114] By combining the PPI-RWR and RGP algorithms, the present invention can identify gene features that are closely related to the efficacy of immunotherapy from multiple perspectives. This not only provides a new idea for existing immunotherapy research, but also provides strong support for future personalized treatment strategies. The biomarkers screened based on this method can be further applied to clinical diagnosis, helping doctors to make more accurate predictions of patients' immune responses before treatment, and then formulate more effective treatment plans.

[0115] (ii) AL-Diffusion data enhancement technology is proposed to effectively amplify small sample data sets.

[0116] Diverse and rich data are the basis for training high-performance machine learning models. However, due to the high difficulty and high cost of immunotherapy clinical trials, and the lack of unified experimental standards, the amount of data currently available is relatively small. Since machine learning is highly dependent on data, the quality and quantity of data directly affect the predictive model's ability to predict unknown samples. The limitations of existing experimental data have become a bottleneck hindering the prediction of immunotherapy efficacy.

[0117] Data augmentation is an effective method to expand the size of data samples. Its core idea is to reduce the risk of overfitting of the model and improve its generalization ability by appropriately modifying existing data or generating new samples based on existing data. Data augmentation technology has achieved remarkable success in the field of computer vision and has become a common means to improve model performance. Unlike data augmentation, which focuses on expanding the size of data, active learning focuses on selecting the most valuable samples to achieve the best model performance with the least data. Active learning evaluates the value of unlabeled data based on the current model, selects high-value samples for expert annotation, reduces the need for a large number of experimental annotations, and improves the predictive performance of the model. Although data augmentation and active learning have developed independently in their respective fields, research combining the two is relatively rare, especially in the field of immunotherapy efficacy prediction, where this combination has not yet been fully applied.

[0118] See also Figure 7In order to solve this problem, this paper introduces generative deep diffusion models (Generative Diffusion Models) from the field of computer vision to immunotherapy research for the first time, and combines active learning to propose an adaptive AL-Diffusion method for achieving higher quality data enhancement. The advantage of the diffusion model is that it can generate new samples with high biological significance by gradually adding and removing noise, effectively simulating the potential distribution of immunotherapy data, and thus generating samples close to real patient data. In image data, the diffusion model generates realistic new images through denoising; in immunotherapy data, it generates new samples that meet biological characteristics by learning the potential distribution of existing patient samples, thereby effectively addressing the challenge of insufficient data.

[0119] The advantage of the diffusion model is that the new samples generated are more continuous and smoothly distributed in the latent space, thus filling the gaps in the number of samples and uneven distribution in clinical trials and improving the model's robustness to samples outside the distribution. In addition, the diffusion model can flexibly capture subtle differences between samples, and the generated samples are not only highly diverse, but also retain important biological feature information, which plays an important role in improving the accuracy of immunotherapy efficacy prediction.

[0120] However, generative models may also produce some samples that deviate from biological significance. In this case, relying solely on data augmentation may not improve the performance of the model, and may even burden the training process and reduce prediction accuracy. To this end, combining active learning can effectively solve this problem. Active learning selects the most valuable samples for training to ensure that the generalization ability of the model is further improved, while avoiding the negative impact of generated data on model performance. Active learning uses adaptive design strategies and combines optimization algorithms to intelligently select the most valuable samples in each round of training based on the prediction results and uncertainties of the model. The AL-Diffusion method cleverly combines active learning and generative diffusion models. Active learning screens and monitors the quality of generated samples to ensure that the generated data is of high quality. At the same time, the diffusion model reduces the dependence of active learning on experimental annotations. The two complement each other and achieve efficient data augmentation.

[0121] This method not only effectively solves the problem of small number of samples in the field of immunotherapy, but also further improves the generalization and extrapolation performance of the model by generating high-quality samples. This innovative data enhancement strategy provides a new solution for the prediction of immunotherapy efficacy, significantly improves the processing ability of small sample data, and is expected to promote the individualization and precision development of immunotherapy.

[0122] (III) A prediction model with superior performance was constructed.

[0123] In the context of predicting the efficacy of immunotherapy, the differences in raw gene expression data are large. Directly using these data for model training may cause highly expressed genes to have an excessive impact on the model, while the effects of low-expressed genes may be weakened or even ignored. In order to reduce this impact, the present invention adopts a min-max normalization method, which scales the expression values ​​of each gene to the same interval to ensure that each gene has the same weight to the objective function of the model and balances the impact of each gene on the prediction results. This processing method not only improves the comparability between features, but also effectively reduces the deviation caused by different feature scales, thereby enhancing the stability of model training and the accuracy of prediction.

[0124] On the basis of completing min-max standardization and AL-Diffusion data enhancement, the present invention also uses grid search and ten-fold cross validation to compare the modeling effects of six different machine learning algorithms. In the end, the XGBoost (XGB) model performed best, with an AUC value of 0.978 on the training set, and was selected as the optimal model. In order to further evaluate the generalization ability of the model, we verified it on the constructed external test set, and the AUC value reached 0.901, proving the good robustness and predictive ability of the model in predicting the efficacy of immunotherapy.

Claims

1. A method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence, characterized in that: The following steps are involved: S100: Analyze the data of gastric cancer patients receiving immunotherapy and construct an immune signature dataset; S200: Write a Python program to screen features based on the PPI-RWR algorithm and the random gene perturbation algorithm; S300: Amplify the sample using a data enhancement algorithm based on a diffusion model; S400: Use six machine learning algorithms to model and predict the standardized data obtained in step S300, and find the optimal parameters of each model through ten-fold cross validation and grid parameter search; S500: Each machine learning algorithm performs ten-fold cross validation under the optimal hyperparameters, compares and selects the optimal machine learning model, and builds the final model; S600: Construct an external test set to verify the model obtained in step S400.

2. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 1, characterized in that: The step S100 specifically includes: Step S110, downloading RNA-seq data of gastric cancer patients who have received immunotherapy from ICBaltas as an initial data set; Step S120, preprocessing the data: using Python Pandas package to implement data filtering, missing value filling, and min-max data standardization; Step S130, using the R package limma, according to the conditions of |log2FC|≥1 and FDR<0.05, screen and retain differentially expressed genes; Step S140: Use Python's Numpy package to perform logarithmic processing on the data.

3. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 1, characterized in that: The step S200 specifically includes: Step S210, constructing a target network, downloading a human PPI network from the STRING database v.12.0; Step S220, selecting existing immunotherapy target genes (PD-1, PD-L1, CTLA-4) and immune-related pathway (T cell receptor, B cell receptor, antigen presentation) genes as seed genes, using the page-rank method of PythonNetworkX, and determining immunotherapy-related genes through the PPI-RWR algorithm; Step S230, iteratively running the random gene perturbation algorithm until the evaluation index begins to decrease, and continuing to screen key biomarkers.

4. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 1, characterized in that: The step S300 specifically includes: Step S310, using the data set D0 obtained in step S200 to construct an agent model C0 for predicting the efficacy of immunotherapy; Step S320: Run the diffusion model to perform data enhancement to obtain X enhanced samples, which constitute the sample pool U0 required for active learning; Step S330: Use the proxy model obtained in step S310 to predict the sample pool obtained in step S320 and calculate the expected improvement value EI of each sample to be selected, sort the expected improvement value of each sample to be selected in descending order, select the top 50 samples and add them to the original data set D0 to form the enhanced data set D1. At this time, there are (X-50) samples left in the sample pool U1 to be selected; Step S340: Retrain the enhanced dataset D1 in step S330 to obtain an updated proxy model C1. Use the proxy models C1 and C1 to predict the sample pool U1 and calculate the expected improvement EI. After sorting in descending order, select the top 50 samples and add them to the dataset D1 to form the enhanced dataset D2. At this time, there are (X-100) samples left in the sample pool U2 to be selected. Step S350: Retrain the enhanced dataset D2 in step S340 to obtain an updated proxy model C2, and make predictions on the sample pool U2 and calculate the expected improvement EI. After sorting in descending order, select the top 50 samples and add them to the dataset D2 to form the enhanced dataset D3. At this time, there are (X-150) samples left in the sample pool U3 to be selected. Step S360: Use the enhanced data set D3 in step S330 to retrain and obtain the updated proxy model C3, make predictions on the sample pool U3, and calculate the expected improvement EI. After sorting in descending order, select the top 50 samples and add them to the data set D3 to form the enhanced data set D4. At this time, there are (X-200) samples left in the sample pool U4 to be selected.

5. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 4, characterized in that: The step S400 specifically includes: Step S410: determining the parameter optimization space of each machine learning algorithm; Step S420, divide the data set according to ten-fold cross validation to ensure that the distribution of each data is as uniform as possible, where 9 data are used as training sets and the remaining 1 data is used as a test set; Step S430: When executing the parameter optimization process, for each set of hyperparameters, first train the model on the training set of step S420, and calculate the evaluation index of the classification model on the test set; the whole process is repeated 10 times to ensure that each piece of data is used as a training set and a test set, and finally obtain the results of 10 sets of evaluation indicators; take the average of these 10 sets of results as the performance of the model under this set of hyperparameters; traverse all the hyperparameter combinations defined in step S410, calculate the evaluation index of each combination, and finally select the set of hyperparameters with the best performance as the optimal hyperparameters; other machine learning algorithms follow the same steps to determine their respective optimal hyperparameters; Step S440: Execute steps S420 and S430 on the enhanced data sets D2, D3, and D4 in sequence to obtain the optimal hyperparameters of each machine learning algorithm under different data sets.

6. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 5, characterized in that: The step S500 specifically includes: Step S510: Based on the enhanced data set D1, ten-fold cross validation is performed in turn for each machine learning algorithm according to the optimal hyperparameters obtained in step S400 to obtain ten average evaluation indicators R2, RMSE and MAE, and the optimal model is selected according to the criteria of maximum R2 and minimum RMSE and MAE; Step S520, executing step S510 on the enhanced data sets D2, D3, and D4 in sequence, and selecting the optimal model corresponding to each data set; Step S530, comparing the model performances under different enhanced data sets obtained in step S510 and step S520, and selecting the optimal enhanced data set and the optimal machine learning algorithm; Step S540: According to the comparison result of step S530, it is determined that the optimal enhanced data set is D4 and the optimal friction sensitivity prediction algorithm is XGB. Therefore, the final immunotherapy efficacy prediction model is constructed based on all the data of D4 using the optimal hyperparameters.

7. The method for pre-evaluation of gastric cancer immunotherapy based on artificial intelligence method according to claim 6, characterized in that: The step S600 specifically includes: S610, gastric cancer patient samples were collected from the GEO database, and the batch effect was first eliminated using the combat package of the R language. Then, the TIDER package of the R language was used to simulate the results of patients receiving immunotherapy, and an external test set was constructed for subsequent verification. S620 , using the optimal XGB model constructed in step S540 to predict the external test set generated in step S610 , and calculating the evaluation index of the model to evaluate the generalization ability of the model.

Images