An assisted chemotherapeutic immunotherapy response method and apparatus for oral squamous cell carcinoma

Abstract

The application relates to an oral squamous cell carcinoma auxiliary chemical immunotherapy response method and device, which comprises the following steps: acquiring image data and baseline clinical factors of a target user, the image data comprising metabolic parameters and image features of a tumor region; performing clinical parameter standardization processing on the metabolic parameters to obtain a standardized metabolic parameter set related to treatment response; performing stability screening on the image features by using an intragroup correlation coefficient to obtain key features related to treatment response; performing regression analysis on the baseline clinical factors to identify independent prognostic factors related to treatment response in the baseline clinical factors; and performing feature fusion on the standardized metabolic parameter set, the key features and the independent prognostic factors to obtain a response rate and a prognosis prediction result of the target user. The method can more accurately and comprehensively predict the response rate and the prognosis of the target user to the auxiliary chemical immunotherapy.

Description

Technical Field

[0001] This invention relates to the field of medical adjuvant technology, specifically to a method and apparatus for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma. Background Technology

[0002] Oral squamous cell carcinoma (OSCC) is the most common type of malignant tumor in the head and neck.

[0003] Currently, while neoadjuvant chemoradiotherapy (NAICT) offers new hope for the treatment of OSCC, its response rate exhibits significant individual variability, with a major pathologic response (MPR) rate of only about 60%. Therefore, existing technologies face the challenge of accurately predicting patient response rates and prognoses to adjuvant chemoradiotherapy. Summary of the Invention

[0004] This invention provides a method and apparatus for adjuvant chemoimmunotherapy response in oral squamous cell carcinoma, aiming to solve the problem that existing technologies have difficulty in accurately predicting the response rate and prognosis of patients to adjuvant chemoimmunotherapy.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A method for responding to adjuvant chemoimmunotherapy in oral squamous cell carcinoma includes:

[0007] Acquire imaging data and baseline clinical factors of the target user, wherein the imaging data includes metabolic parameters and radiomics features of the tumor region;

[0008] The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0009] Stability screening of the radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response.

[0010] Regression analysis was performed on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors;

[0011] The standardized metabolic parameter set, the key features, and the independent prognostic factors are fused to obtain the response rate and prognostic prediction results for the target user.

[0012] Optionally, the radiomics features include shape features, first-order statistical features, and texture features; obtaining the radiomics features of the target user includes:

[0013] The image data of the target user is subjected to data desensitization and de-identification, format unification, normalization and image quality check to obtain preprocessed image data;

[0014] The preprocessed image data is resampled to obtain isotropic image data;

[0015] Feature extraction is performed on the isotropic image data to obtain shape features, first-order statistical features, and texture features.

[0016] Optionally, the metabolic parameters include maximum standardized uptake, average standardized uptake, peak standardized uptake, metabolic tumor volume, total glycolysis damage, target-to-background ratio, tumor volume, and total lesion activity; the standardization of the metabolic parameters to obtain a set of standardized metabolic parameters related to treatment response includes:

[0017] The metabolic parameters and their changes are processed using standard fractions to obtain standardized metabolic parameters.

[0018] The standardized metabolic parameters were screened using intragroup correlation coefficients to obtain a set of standardized metabolic parameters related to treatment response.

[0019] Optionally, the stability screening of the radiomics features using intragroup correlation coefficients to obtain key features related to treatment response includes:

[0020] Acquire two sets of image data of the target user at two time points within a preset time interval;

[0021] The radiomics features of the two sets of image data were extracted separately, and the intra-group correlation coefficient of each radiomics feature was calculated.

[0022] Radiomic features with intragroup correlation coefficients greater than a preset threshold were identified as key features associated with treatment response.

[0023] Optionally, the baseline clinical factors include age, sex, maximum standardized uptake, metabolic tumor volume, total glycolytic damage, TNM stage, histological differentiation, T stage, N stage, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and body mass index; the regression analysis of the baseline clinical factors identifies independent prognostic factors related to treatment response, including:

[0024] Univariate regression analysis was performed on each baseline clinical factor to obtain the first baseline clinical factor associated with survival outcome;

[0025] Multivariate regression analysis was performed on the first baseline clinical factors to obtain the second baseline clinical factors associated with survival outcomes.

[0026] The residual variable test was performed on the second baseline clinical factor to determine the second baseline clinical factor that satisfies the proportional risk assumption as the independent prognostic factor.

[0027] Optionally, the step of fusing the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the response rate of the target user includes:

[0028] The standardized metabolic parameter set and the key features are weighted and fused to obtain a radiomics score;

[0029] The radiomics score and the independent prognostic factors are then re-fused to obtain the response rate of the target user.

[0030] Optionally, the step of fusing the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the prognostic prediction result for the target user includes:

[0031] The standardized metabolic parameter set and the key features are weighted and fused to obtain a radiomics score;

[0032] The radiomics score and the independent prognostic factors are fused using proportional hazards, and the fused hazard score that satisfies the proportional hazards assumption is determined as the prognostic prediction result for the target user.

[0033] A device for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma, comprising:

[0034] The data acquisition module is used to acquire the target user's imaging data and baseline clinical factors, including metabolic parameters and radiomics features of the tumor region.

[0035] The standardization processing module is used to perform clinical parameter standardization processing on the metabolic parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0036] The stability screening module is used to screen the radiomics features for stability using intragroup correlation coefficients to obtain key features related to treatment response.

[0037] The regression analysis module is used to perform regression analysis on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors;

[0038] The treatment response module is used to perform feature fusion of the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the response rate and prognostic prediction results of the target user.

[0039] An electronic device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps:

[0040] Acquire imaging data and baseline clinical factors of the target user, wherein the imaging data includes metabolic parameters and radiomics features of the tumor region;

[0041] The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0042] Stability screening of the radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response.

[0043] Regression analysis was performed on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors;

[0044] The standardized metabolic parameter set, the key features, and the independent prognostic factors are fused to obtain the response rate and prognostic prediction results for the target user.

[0045] A computer-readable storage medium having a computer program stored thereon, the computer program being loaded by a processor to perform steps in a method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma.

[0046] In this embodiment, by integrating multi-dimensional information from imaging data (metabolic parameters, radiomics features) and baseline clinical factors, metabolic parameters are standardized to reduce variability and enhance their correlation with treatment response. Consistent and reliable radiomics key features are obtained through stability screening. Independent prognostic factors in baseline clinical factors are identified through regression analysis, and then multi-source features are fused for prediction. This enables a more accurate and comprehensive prediction of the target user's response rate and prognosis to adjuvant chemoimmunotherapy, providing a precise basis for the clinical development of personalized treatment plans. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in this embodiment, the accompanying drawings used in the description of the embodiment will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A schematic flowchart illustrating an embodiment of the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma provided by the present invention;

[0049] Figure 2A schematic diagram of the research framework for an embodiment of the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma provided by the present invention;

[0050] Figure 3 A schematic diagram of an embodiment of the adjuvant chemoimmunotherapy response device for oral squamous cell carcinoma provided by the present invention;

[0051] Figure 4 This is a schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0052] The technical solutions in this embodiment will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] In the following description, specific embodiments of the invention will be illustrated with reference to steps and symbols performed by one or more computers, unless otherwise stated. Therefore, these steps and operations will be referred to several times as being performed by a computer, and computer execution as referred to herein includes operations by a computer processing unit representing electronic signals of data in a structured format. This operation transforms the data or maintains it at a location in the computer's memory system, which can be reconfigured or otherwise alter the operation of the computer in a manner well known to those skilled in the art. The data structure maintained by the data is the physical location of the memory, which has specific characteristics defined by the data format. However, the principles of the invention described above are not intended to be limiting, and those skilled in the art will understand that many of the steps and operations described below can also be implemented in hardware.

[0054] The terms "module" or "unit" as used herein can be considered as software objects executing on the computing system. The different components, modules, engines, and services described herein can be considered as implementation objects on the computing system. The apparatus and methods described herein are preferably implemented in software, but can also be implemented in hardware, both of which are within the scope of this invention.

[0055] This invention provides a method and apparatus for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma. Please refer to [link to relevant documentation]. Figure 1 , Figure 1 A schematic flowchart of an embodiment of the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma provided by the present invention includes:

[0056] S101: Acquire imaging data and baseline clinical factors of the target user. The imaging data includes metabolic parameters and radiomics characteristics of the tumor region.

[0057] In one specific embodiment, the target users mainly include three groups: first, patients with oral squamous cell carcinoma (OSCC) who have undergone surgical resection or histopathological diagnosis and are scheduled for neoadjuvant chemoimmunotherapy, who present with complaints such as persistent oral ulcers, leukoplakia or lumps, and pain; second, OSCC patients as an external validation cohort to evaluate the stability and reproducibility of the model; and third, healthy subjects matched for gender and age with the patient group as controls.

[0058] Imaging data refers to 18F-FDG and 18F-FAPI imaging data acquired through multimodal PET / CT technology, including metabolic parameters and radiomics characteristics of the tumor region.

[0059] Baseline clinical factors refer to clinical and imaging indicators related to patients, including age, sex, maximum standard uptake (SUVmax), metabolic tumor volume (MTV), total damaged glycolysis (TLG), TNM stage, histological differentiation, T stage, N stage, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and body mass index (BMI).

[0060] Metabolic parameters are key indicators for quantifying tumor metabolic activity, including maximum standardized uptake (SUVmax), mean standardized uptake (SUVmean), peak standardized uptake (SUVpeak), metabolic tumor volume (MTV), total damaged glycolysis (TLG), target-to-background ratio (TBR), tumor volume (TV), and total lesion activity (TLA).

[0061] Radiomics features are quantitative features extracted from image data, including shape features, first-order statistical features, and texture features. Shape features describe the geometric morphology of the tumor (such as surface area and volume); first-order statistical features reflect the distribution pattern of pixel gray levels (such as mean and entropy); texture features capture the complexity of pixel spatial distribution through methods such as gray-level region length matrix (GLSZM), neighborhood gray-level difference matrix (NGTDM), gray-level run length matrix (GLRLM), and gray-level co-occurrence matrix (GLCM), such as region size entropy and short run low gray-level emphasis. These features can be used to reveal the heterogeneity of tumor microstructure.

[0062] S102: Perform clinical parameter standardization on metabolic parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0063] In one specific embodiment, the standardized metabolic parameter set is a set of parameters related to treatment response obtained by standardizing tumor metabolic parameters according to clinical parameters.

[0064] S103: Stability screening of radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response;

[0065] In one specific embodiment, the key feature is the treatment response-related feature obtained after stability screening of radiomics features through intragroup correlation coefficients.

[0066] S104: Perform regression analysis on baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors;

[0067] In one specific embodiment, independent prognostic factors refer to baseline clinical factors that are significantly associated with the survival outcomes (overall survival (OS) and progression-free survival (PFS)) of patients with oral squamous cell carcinoma, as identified through regression analysis.

[0068] S105: The standardized set of metabolic parameters, key features, and independent prognostic factors are fused to obtain the response rate and prognostic prediction results for the target users.

[0069] In one specific embodiment, the response rate refers to the proportion of patients with oral squamous cell carcinoma who achieve a significant pathological response (MPR) to neoadjuvant chemoimmunotherapy (NAICT).

[0070] Prognostic prediction results refer to the risk assessment results of patients' OS and PFS based on a composite model (radiomics score + independent prognostic factors). It can quantify the predicted survival rate at different time points (1 year, 2 years, 3 years) and provide a basis for individualized treatment decisions.

[0071] In this embodiment, by integrating multi-dimensional information from imaging data (metabolic parameters, radiomics features) and baseline clinical factors, metabolic parameters are standardized to reduce variability and enhance their correlation with treatment response. Consistent and reliable radiomics key features are obtained through stability screening. Independent prognostic factors in baseline clinical factors are identified through regression analysis, and then multi-source features are fused for prediction. This enables a more accurate and comprehensive prediction of the target user's response rate and prognosis to adjuvant chemoimmunotherapy, providing a precise basis for the clinical development of personalized treatment plans.

[0072] In one specific embodiment, in S101, the radiomics features include shape features, first-order statistical features, and texture features; obtaining the radiomics features of the target user includes: performing data desensitization and de-identification, format unification, normalization, and image quality checks on the image data of the target user to obtain preprocessed image data; resampling the preprocessed image data to obtain isotropic image data; and extracting features from the isotropic image data to obtain shape features, first-order statistical features, and texture features.

[0073] It should be noted that shape features are an important component of radiomics features, mainly used to describe the geometric morphological attributes of tumors, including spatial morphological parameters such as tumor size, surface area, volume, and compactness. By quantifying the external features of tumors, basic morphological information is provided for subsequent analysis.

[0074] First-order statistical features are features that reflect the distribution pattern of pixel gray values ​​in an image. By calculating statistical parameters such as the mean, median, standard deviation, and entropy of gray values, they describe the overall distribution characteristics of gray values ​​within a tumor. They do not involve the spatial relationship between pixels, but only reflect the statistical distribution characteristics of gray values.

[0075] Texture features are extracted using methods such as gray-scale region length matrix (GLSZM), neighborhood gray-scale difference matrix (NGTDM), gray-scale run length matrix (GLRLM), and gray-scale co-occurrence matrix (GLCM). These features are used to capture the spatial distribution patterns and complexity of pixel gray levels, such as region size entropy and short run length low gray-scale emphasis, which can reflect the heterogeneity and spatial distribution patterns of tumor microstructure.

[0076] In one specific embodiment, the imaging feature analysis based on dual-parameter 18F-FDG and 18F-FAPI begins with a preprocessing step, including data desensitization and de-identification, format standardization, normalization, and image quality checks to ensure that the acquired image data meets the analysis standards. Then, the scan results are read and interpreted according to a unified standardized process, extracting key imaging features such as tumor size, shape, margins, internal signals, and invasion of adjacent tissues.

[0077] In practice, two experienced nuclear medicine physicians were selected to interpret all subject images, and a third party was involved in the discussion in case of disputes. Lesions were classified as primary tumors, lymph node metastases, or distant metastases. Increased uptake of 18F-FDG and 18F-FAPI was considered indicative of tumor location. The uptake of imaging agents in tumor lesions was quantified using the maximum standardized uptake value (SUVmax) and the target-to-background ratio (TBR). Regions of interest (ROIs) were delineated and segmented using a 40% SUVmax threshold, avoiding necrotic and cystic areas during calculation. The number and size (long axis, cm) of lesions were recorded. Normal non-tumor tissue surrounding the lesions was selected to delineate ROIs, and their mean standardized uptake value (SUVmean) was calculated.

[0078] The formula for calculating TBR is: TBR = SUVmax / SUVmean. Further measurements are taken using PET / CT to determine the metabolic tumor volume (MTV), mean standardized uptake value (SUVmean), and total lesion glycolysis (TLG), where TLG = MTV × SUVmean.

[0079] Before extracting features from radiomics, image preprocessing is required. This ensures the rotation invariance of the extracted texture features and the comparability of data from images acquired from different batches. First, the images need to be resampled to make the resolution consistent in the X, Y, and Z directions, thus converting anisotropic PET images into isotropic ones. In this study, all image voxels were resampled to a size of 2mm x 2mm x 2mm. Furthermore, according to the requirements of LIFEx software, the minimum voxel size of VOI is 64 voxels for further study, which is a prerequisite for further analysis. Ultimately, the software automatically extracted and calculated 124 PET image omics features for each case. The extracted features all conformed to the International Biomarker Standardization Initiative (IBSI) and included the following types of features: shape features, first-order statistical features, and texture features including the Grey Level SizeZone Matrix (GLSZM), Neighborhood Grey Tone Difference Matrix (NGTDM), Grey Level Run Length Matrix (GLRLM), and Grey Level Cooccurrence Matrix (GLCM).

[0080] In this embodiment, standardized preprocessing (data anonymization, format unification, normalization, and quality checks) and isotropic resampling ensured the comparability of image data and the rotation invariance of texture features. Combined with experienced physician interpretation and a third-party dispute resolution mechanism, the accuracy of lesion identification was improved. By delineating ROIs that avoid necrotic and cystic areas and calculating multi-dimensional quantitative indicators such as SUVmax, TBR, MTV, and TLG, a comprehensive characterization of tumor morphology and metabolic features was achieved. Automatic extraction of 124 radiomics features conforming to IBSI standards improved feature extraction efficiency and enhanced the reliability and universality of the features, providing high-quality quantitative evidence for subsequent tumor diagnosis, prognostic assessment, and other analyses.

[0081] In one specific embodiment, in S102, the metabolic parameters include maximum standardized uptake, average standardized uptake, peak standardized uptake, metabolic tumor volume, total glycolysis damage, target-to-background ratio, tumor volume, and total lesion activity. The metabolic parameters are then subjected to clinical parameter standardization to obtain a set of standardized metabolic parameters related to treatment response. This includes: standardizing the metabolic parameters and their variations to obtain standardized metabolic parameters; and screening the standardized metabolic parameters using intragroup correlation coefficients to obtain the set of standardized metabolic parameters related to treatment response.

[0082] It should be noted that the maximum standardized uptake value (SUVmax) is the highest value reflecting the local uptake of radiotracers in a tumor and is one of the core quantitative indicators for assessing tumor metabolic activity.

[0083] The Mean Standardized Uptake Value (SUVmean) is the average radiotracer uptake of all pixels within the region of interest (ROI) of the tumor, reflecting the overall metabolic level of the tumor. It is often used in conjunction with metabolic tumor volume (MTV) to calculate total damaged glycolysis (TLG).

[0084] Peak Standardized Uptake Value (SUVpeak) is the average uptake value of a fixed small volume region within a tumor (usually a sphere of 1 cm³ or 10 mm in diameter). It can reduce the influence of part of the volume effect and more stably reflect the metabolic activity of the tumor.

[0085] Metabolic tumor volume (MTV) is the volume of a metabolically active region of a tumor delineated by a preset threshold (such as 40% SUVmax). It represents the spatial distribution range of tumor metabolic activity and is an important parameter for assessing tumor burden.

[0086] Total Lesion Glycolysis (TLG) is calculated by multiplying the metabolic tumor volume (MTV) and the mean standard uptake value (SUVmean) (TLG = MTV × SUVmean), and comprehensively reflects the total metabolic load of the tumor.

[0087] The target-to-background ratio (TBR) is the ratio of the maximum standard uptake value (SUVmax) of the tumor region to the average standard uptake value (SUVmean) of the surrounding normal tissue. It is used to enhance the signal contrast between tumors and normal tissues and improve the specificity of lesion detection.

[0088] Tumor volume (TV) is the anatomical volume of a tumor based on image delineation. It reflects the physical size of the tumor and, when combined with metabolic parameters, can assess the structural and metabolic characteristics of the tumor.

[0089] Total Lesion Activity (TLA) is the total uptake of radiotracers within a tumor region. It is usually calculated by multiplying the tumor volume (TV) by the average uptake value and is used to quantify the overall metabolic activity of the tumor.

[0090] The intra-rater correlation coefficient (ICC) is a statistical indicator used to assess the consistency and reliability of the measurement results of the same evaluator on the same object at different times or under different conditions.

[0091] In one specific embodiment, in S103, the stability of radiomics features is screened by intragroup correlation coefficient to obtain key features related to treatment response, including: acquiring two sets of image data of the target user at two time points within a preset time interval; extracting radiomics features of the two sets of image data respectively, and calculating the intragroup correlation coefficient of each radiomics feature; determining the radiomics features with intragroup correlation coefficients greater than a preset coefficient threshold as key features related to treatment response.

[0092] The preset time interval can be 2 weeks or other time intervals. The specific time interval can be flexibly adjusted according to clinical needs and the stability of imaging equipment, and is not limited here.

[0093] In one specific embodiment, in S104, baseline clinical factors include age, sex, maximum standardized uptake, metabolic tumor volume, total glycolytic damage, TNM stage, histological differentiation, T stage, N stage, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and body mass index. Regression analysis is performed on the baseline clinical factors to identify independent prognostic factors related to treatment response, including: univariate regression analysis on each baseline clinical factor to obtain a first baseline clinical factor related to survival outcome; multivariate regression analysis on the first baseline clinical factor to obtain a second baseline clinical factor related to survival outcome; and residual variable testing is performed on the second baseline clinical factor to determine that the second baseline clinical factor satisfying the proportional risk assumption is an independent prognostic factor.

[0094] Age refers to the patient's actual age and is used as one of the baseline clinical factors for intergroup matching and correction of confounding factors, such as removing age as a covariate in radiomics feature analysis.

[0095] Gender / Sex refers to the patient's biological sex (male / female) and is used as one of the baseline clinical factors for intergroup matching and correction of confounding factors, such as removing gender as a covariate in radiomics feature analysis.

[0096] The maximum standardized uptake value (SUVmax) is the highest value reflecting the local uptake of radiotracers in a tumor and is a core quantitative indicator for assessing tumor metabolic activity.

[0097] Metabolic tumor volume (MTV) is the volume of a metabolically active region of a tumor, delineated by a preset threshold (such as 40% SUVmax). It represents the spatial distribution range of tumor metabolic activity and is an important parameter for assessing tumor burden.

[0098] Total Lesion Glycolysis (TLG) is calculated by multiplying the metabolic tumor volume (MTV) and the mean standard uptake value (SUVmean) (TLG = MTV × SUVmean), and comprehensively reflects the total metabolic load of the tumor.

[0099] TNM staging is a staging system based on the primary tumor site (T), regional lymph nodes (N), and distant metastases (M) to assess the degree of tumor progression and prognosis.

[0100] Histological differentiation describes the degree of similarity between tumor cells and normal cells. It is usually classified as well-differentiated, moderately differentiated, and poorly differentiated. The lower the degree of differentiation, the higher the malignancy of the tumor, which is an important factor affecting prognosis.

[0101] T stage (Tumor Stage) is an indicator in the TNM staging system that describes the size and extent of invasion of the primary tumor, reflecting the degree of local tumor progression, such as tumor diameter and whether it has invaded adjacent tissues.

[0102] N-stage (Node Stage) is an indicator in the TNM staging system that describes regional lymph node metastasis, reflecting the extent of lymph node involvement in the tumor, such as lymph node size, number, and location of metastasis.

[0103] The neutrophil-to-lymphocyte ratio (NLR) is the ratio of the neutrophil count to the lymphocyte count in peripheral blood. It reflects the body's inflammatory state and immune function and is a potential biomarker for assessing the prognosis of cancer patients.

[0104] The platelet-to-lymphocyte ratio (PLR) is the ratio of the platelet count to the lymphocyte count in peripheral blood. It reflects the body's coagulation function and immune status and is associated with tumor progression and prognosis.

[0105] Body Mass Index (BMI) is the ratio of weight (kg) to the square of height (m) (BMI = weight / height²). It is used to assess the degree of obesity in a person and is one of the clinical factors that affect the occurrence, development, and treatment response of tumors.

[0106] Univariate regression analysis is a statistical method used to analyze the association between each baseline clinical factor and survival outcome individually, and to screen for potential factors that may affect prognosis. Specifically, in this application, univariate Cox regression analysis was used to test baseline clinical factors such as age, sex, and SUVmax one by one, and factors with P < 0.05 were included in subsequent multivariate analysis.

[0107] Survival outcome refers to the patient's survival status or disease progression outcome assessed during follow-up. Specifically, the survival outcome in this application mainly includes overall survival (OS) and progression-free survival (PFS). OS is defined as the time from surgical pathological diagnosis to death or the last follow-up, and PFS is defined as the time from diagnosis to disease progression, death, or the end of follow-up.

[0108] Multivariate regression analysis, based on univariate analysis, incorporates multiple potential influencing factors and uses statistical models (such as Cox regression) to assess the independent impact of each factor on survival outcomes, screening out statistically significant independent prognostic factors. Specifically, the multivariate Cox regression in this application is used to identify key clinical variables (such as N stage and histological differentiation) related to OS and PFS.

[0109] Residual variable testing is a method used to verify whether the assumptions of a multifactor regression model (such as the Cox model) are met. Specifically, this application uses the Schoenfeld residual method to test whether the variables meet the proportional risk assumption. If the test result P > 0.05, it indicates that the variable meets the assumption and can be included in the model; if P ≤ 0.05, the variable needs to be stratified or time-dependent.

[0110] It should be noted that, in order to avoid overfitting during the model building process, the feature selection process is as follows: (1) Two weeks after the initial feature extraction, PET images of 50 patients were randomly selected for ROI delineation and feature extraction again. Subsequently, intra-rater correlation coefficient (ICC) analysis was performed to evaluate the consistency of the features and assess the reliability and repeatability of the extracted features. Features with an ICC value greater than 0.9 were considered to have good robustness and were retained. (2) Due to the large differences in the dimensions of different features, in order to eliminate the differences caused by the dimensions and make different features comparable, we applied z-score standardization to further process the data. (3) Common feature selection methods include filtering, wrapping, and embedding. Generally, the embedded algorithm is preferred, namely the Least Absolute Shrinkage and Selection algorithm. The LASSO Operator (LASSO) algorithm, in conjunction with the Cox model, filters features. This algorithm reduces redundant features, achieves feature dimensionality reduction, and selects the best features for predicting OS and PFS.

[0111] Specifically, by re-delineating ROIs and extracting features from the same batch of patients' image data two weeks after the initial feature extraction, the ICC values ​​of the extracted features were calculated to evaluate the stability and repeatability of radiomics features. Features with ICC values ​​greater than 0.9 were considered to have good robustness and were retained for subsequent analysis.

[0112] The radiomics features selected using the LASSO algorithm were weighted and calculated with their respective LASSO coefficients to determine the PET radiomics score for each patient. Then, based on clinical and radiological information including age, sex, SUVmax, MTV, TLG, TNM stage, histological differentiation, T stage, N stage, NLR, PLR, and BMI, univariate and multivariate Cox regression analyses were used in the training group to screen for independent prognostic factors related to OS and PFS. SUVmax, MTV, TLG, NLR, and PLR were grouped by their median for analysis. In the univariate Cox regression, factors with P < 0.05 were included in the multivariate Cox regression analysis. The Schoenfeld residual method was used to test whether the variables included in the multivariate Cox regression analysis met the proportional risk assumption (P > 0.05). If the assumption was met, the next step of analysis was performed. Based on the results of the multivariate analysis, a clinical model was constructed according to clinically relevant variables. This clinical model was then combined with the PET radiomics score to construct a composite model.

[0113] To quantify the performance of the prognostic models and compare them with the traditional TNM staging system, the C-index was used to determine the discriminative power of each model in both the training and validation groups. Time-dependent C-index curves were used in both groups to evaluate the prognostic performance of each model at different time points. The model with the best discriminative power in predicting overall survival (OS) and progression-free survival (PFS) was used to construct the corresponding nomogram, a visual model that can intuitively assess the risk probability of each patient. ROC curves were plotted in both the training and validation groups, and the discriminative power of the nomogram was further evaluated based on the area under the curve (AUC) of the 1-year, 2-year, and 3-year ROC curves. The predictive accuracy of the nomogram was evaluated by plotting calibration curves (1-year, 2-year, and 3-year predicted survival rates) in both the training and validation groups. These curves were analyzed based on the Hosmer-Lemeshow goodness-of-fit test to assess the closeness between the predicted and actual survival rates. Using decision curve analysis (DCA) to evaluate the clinical applicability of prognostic models can help clinicians intervene in high-risk patients and avoid intervention in low-risk patients, thus preventing overtreatment.

[0114] In this embodiment, stable and reliable radiomics features were screened out by intragroup correlation coefficient (ICC>0.9), and baseline independent clinical prognostic factors were identified by univariate / multivariate Cox regression analysis. The clinical-radiomics composite model constructed after dimensionality reduction optimization by LASSO combined with the Cox model showed superior discrimination and predictive accuracy compared to traditional TNM staging through C-index, time-dependent ROC curves (1 / 2 / 3-year AUC), and calibration curves (Hosmer-Lemeshow test). Decision curve analysis (DCA) verified its clinical applicability. The risk of overall survival (OS) and progression-free survival (PFS) can be intuitively and accurately assessed through nomograms, providing a reliable basis for personalized clinical intervention.

[0115] In one specific embodiment, in S105, the standardized metabolic parameter set, key features, and independent prognostic factors are fused to obtain the response rate of the target user, including: weighted fusion of the standardized metabolic parameter set and key features to obtain a radiomics score; and re-fusion of the radiomics score and independent prognostic factors to obtain the response rate of the target user.

[0116] The standardized metabolic parameter set, key features, and independent prognostic factors are fused to obtain the prognostic prediction results for the target user. This includes: weighted fusion of the standardized metabolic parameter set and key features to obtain a radiomics score; proportional hazards fusion of the radiomics score and independent prognostic factors, and determining the fusion hazard score that meets the proportional hazards assumption as the prognostic prediction result for the target user.

[0117] It should be noted that the obtained multimodal 18F-FDG / 18F-FAPI-04 PET / CT metabolic parameters, specifically including SUVmax, peak standardized uptake (SUVpeak), mean standardized uptake (SUVmean), and total lesion glycolysis (TLG) for the primary OSCC lesion and lymph node metastases, were analyzed using the following methods:

[0118] 1. Based on the response to neoadjuvant chemoimmunotherapy (¹ 8 F-FDG metabolic imaging / ¹ 8 Comparison of F-FAPI interstitial imaging features between groups: Two-sample t-tests were performed using the Statistical Analysis function of DPASFA 4.0. Age and sex were removed as covariates. After Gaussian random field (GRF) correction, P < 0.05 was considered statistically significant, and statistically significant metabolic features and t-values ​​were obtained.

[0119] 2,¹ 8 F-FDG metabolic imaging / ¹ 8 Intergroup comparison of F-FAPI interstitial imaging parameters: A general linear model was used, with factors such as subject age, disease duration, stage, and histological grade included as covariates to compare whether there were significant differences in metabolic parameters (SUVmax, SUVmean, SUVpeak, FDG-TV, FDG-TLA, FAPI-TV, FAPI-TLA) between groups.

[0120] 3,¹ 8 F-FDG metabolic imaging / ¹ 8F-FAPI interstitial imaging efficacy assessment: The Mann-Whitney U test was used to compare whether the differences in the primary lesion PET metabolic parameters SUVmax1, SUVmean1, SUVpeak1, TBR1, FDG-TV, FDG-TLA, FAPI-TV1, FAPI-TLA1, SUVmax2, SUVmean2, SUVpeak2, FAPI-TV2, FAPI-TLA2, TBR2, ∆SUVmax, ∆SUVmean, ∆SUVpeak, ∆FAPI-TV, and ∆FAPI-TLA between the treatment response group and the non-responder group were statistically significant.

[0121] 4. Analyze ROC curves and Youden index to determine the optimal cutoff values ​​for statistically significant PET metabolic parameters, and compare their predictive ability for treatment response using AUC, specificity, sensitivity, NPV, and PPV. Compare the differences in lymph node multimodal PET metabolic parameters SUVmax1, SUVmean1, SUVpeak1, FAPI-TV1, FAPI-TLA1, SUVmax2, SUVmean2, SUVpeak2, FDG-TV2, FDG-TLA2, FAPI-TV2, FAPI-TLA2, ∆SUVmax, ∆SUVmean, ∆SUVpeak, ∆FAPI-TV, and ∆FAPI-TLA between the lymph node treatment response and non-response groups, and evaluate their predictive ability for lymph node treatment response.

[0122] 5. Construction of a Neoadjuvant Chemoimmunotherapy Prediction Model for OSCC Based on Artificial Intelligence Algorithms: After ICC analysis, omics features were selected for further research. To avoid model overfitting, we used the LASSO algorithm combined with Cox regression analysis for further feature selection, ultimately identifying features significantly associated with OS and PFS.

[0123] The PET radiomics score for OS is calculated using the following formula:

[0124] PET radiomics score (OS) = 0.314803809 × GLSZM_ZoneSizeEntropy + 0.155396788 × MORPHOLOGICAL_SurfaceArea - 0.002315593 × GLRLM_ShortRunLowGreyLevelEmphasis.

[0125] The PET radiomics score for PFS can be calculated using the following formula:

[0126] PET radiomics score (PFS) = 0.294117864 × GLSZM_ZoneSizeEntropy + 0.033922475 × Morphological_Compacity - 0.158687904 × GLRLM_ShortRunLowGreyLevelEmphasis.

[0127] Based on the results of multivariate Cox regression analysis, the selected clinical factors were used to establish a clinical model. The clinical factors selected based on overall survival (OS) were N-stage and those selected based on progression-free survival (PFS) were N-stage and histological differentiation. The clinical model, combined with PET radiomics scores, constituted the composite model of this study, which was compared with the traditional TNM staging system. Furthermore, a C-index was constructed to represent the prognostic model associated with OS and PFS.

[0128] In this embodiment, multimodal 18F-FDG / 18F-FAPI-04 PET / CT metabolic parameter analysis effectively identified statistically significant metabolic characteristics between the response and non-response groups of neoadjuvant chemoimmunotherapy for OSCC. The LASSO algorithm combined with Cox regression was used to screen radiomics features significantly associated with OS and PFS (avoiding model overfitting). A composite model combining OS and PFS radiomics scores with clinical factors (N stage, histological differentiation) and radiomics scores was constructed. Compared to traditional TNM staging, this model significantly improved the accuracy (such as specificity and sensitivity) of treatment response rate prediction and the performance of prognostic (OS, PFS) assessment (such as a higher C-index).

[0129] In one specific embodiment, please refer to Figure 2 , Figure 2This is a schematic diagram of the research framework for an embodiment of the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma provided by the present invention. The overall framework is divided into five modules: clinical problem, sample collection, data processing, model construction, and clinical translation. The research stems from the clinical challenges of limited improvement in the efficacy of OSCC treatment, including high mortality and DALY rates, individual variability in response rates, low significant pathological remission rates, and difficulties in preoperative assessment and response prediction (such as limitations of traditional imaging, unclear molecular imaging mechanisms of treatment resistance, and unclear changes in the tumor microenvironment during treatment). In the sample collection phase, 60 OSCC patients and a 30-exception validation cohort were included. Imaging data were acquired through dual-probe (18F-FDG and 18F-FAPI) PET / CT scans, and post-processing was performed on indicators such as SUVmax, MTV, and TLG. The data processing phase included standardized preprocessing, three-dimensional resampling, Gaussian filtering for noise reduction, and extraction of shape and texture features. Optimal features were then selected using algorithms such as LASSO and mRMR. Model construction included LASSO-based PET radiomics scoring, univariate / multivariate Cox regression analysis (including clinical and radiomics factors with P < 0.05), and the construction of a composite model combining the clinical model and radiomics scores. The ultimate clinical translation goal is to develop a personalized efficacy prediction model (target AUC ≥ 0.90), evaluate its practicality through decision curve analysis (DCA), and solidify the model by plotting nomograms to distinguish between the best response group and the non-response group.

[0130] To facilitate better implementation of the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma provided by this invention, this invention also provides an apparatus based on the aforementioned adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma. The meanings of the terms used are the same as in the aforementioned adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma, and specific implementation details can be found in the descriptions of the method embodiments.

[0131] Please see Figure 3 , Figure 3 This is a schematic diagram of an embodiment of the adjuvant chemoimmunotherapy response device for oral squamous cell carcinoma provided by the present invention. The adjuvant chemoimmunotherapy response device 500 for oral squamous cell carcinoma may include:

[0132] The data acquisition module 301 is used to acquire the target user's imaging data and baseline clinical factors. The imaging data includes metabolic parameters and radiomics characteristics of the tumor region.

[0133] The standardization processing module 302 is used to perform clinical parameter standardization processing on metabolic parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0134] Stability screening module 303 is used to screen radiomics features for stability through intragroup correlation coefficients to obtain key features related to treatment response.

[0135] The regression analysis module 304 is used to perform regression analysis on baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors;

[0136] The treatment response module 305 is used to fuse standardized metabolic parameter sets, key features and independent prognostic factors to obtain the response rate and prognostic prediction results of the target user.

[0137] The present invention also provides an electronic device, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of an embodiment of the electronic device provided by the present invention, specifically:

[0138] The electronic device may include components such as a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0139] The processor 401 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 402, and by calling data stored in the memory 402, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor. The application processor mainly handles the operation of the storage medium, user interface, and application programs, while the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 401.

[0140] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. The program storage area may store applications required for operating the storage medium and at least one function (such as sound playback function, image playback function, etc.); the data storage area may store data created according to the use of the electronic device. In addition, the memory 402 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0141] The electronic device also includes a power supply 403 that supplies power to various components. Preferably, the power supply 403 can be logically connected to the processor 401 via a power management storage medium, thereby enabling functions such as charging, discharging, and power consumption management through the power management storage medium. The power supply 403 may also include one or more DC or AC power supplies, recharge storage media, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0142] The electronic device may also include an input unit 404, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0143] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 401 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 402 according to the following instructions, and the processor 401 runs the applications stored in the memory 402 to realize various functions, as follows:

[0144] Acquire imaging data and baseline clinical factors of the target users. The imaging data includes metabolic parameters and radiomics characteristics of the tumor region.

[0145] The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0146] Stability screening of radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response.

[0147] Regression analysis was performed on baseline clinical factors to identify independent prognostic factors associated with treatment response among the baseline clinical factors;

[0148] By fusing standardized metabolic parameter sets, key features, and independent prognostic factors, the response rate and prognostic prediction results for the target users are obtained.

[0149] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0150] Therefore, the present invention provides a computer-readable storage medium storing a computer program thereon, the computer program being loaded by a processor to execute the steps in any of the adjuvant chemoimmunotherapy response methods for oral squamous cell carcinoma provided by the present invention. For example, the computer program, when loaded by a processor, can execute the following steps:

[0151] Acquire imaging data and baseline clinical factors of the target users. The imaging data includes metabolic parameters and radiomics characteristics of the tumor region.

[0152] The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response;

[0153] Stability screening of radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response.

[0154] Regression analysis was performed on baseline clinical factors to identify independent prognostic factors associated with treatment response among the baseline clinical factors;

[0155] By fusing standardized metabolic parameter sets, key features, and independent prognostic factors, the response rate and prognostic prediction results for the target users are obtained.

[0156] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0157] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0158] Since the computer program stored in the computer-readable storage medium can execute the steps in any of the adjuvant chemoimmunotherapy response methods for oral squamous cell carcinoma provided by the present invention, the beneficial effects that any of the adjuvant chemoimmunotherapy response methods for oral squamous cell carcinoma provided by the present invention can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0159] The above provides a detailed description of the method and apparatus for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for responding to adjuvant chemoimmunotherapy for oral squamous cell carcinoma, characterized in that, include: Acquire imaging data and baseline clinical factors of the target user, wherein the imaging data includes metabolic parameters and radiomics features of the tumor region; The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response; Stability screening of the radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response. Regression analysis was performed on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors; The standardized metabolic parameter set, the key features, and the independent prognostic factors are fused to obtain the response rate and prognostic prediction results for the target user.

2. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The image omics features include shape features, first-order statistical features, and texture features; Obtain the radiomics features of the target user, including: The image data of the target user is subjected to data desensitization and de-identification, format unification, normalization and image quality check to obtain preprocessed image data; The preprocessed image data is resampled to obtain isotropic image data; Feature extraction is performed on the isotropic image data to obtain shape features, first-order statistical features, and texture features.

3. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The metabolic parameters include maximum standardized uptake, average standardized uptake, peak standardized uptake, metabolic tumor volume, total glycolysis damage, target-to-background ratio, tumor volume, and total lesion activity; the metabolic parameters are then subjected to clinical parameter standardization to obtain a standardized set of metabolic parameters related to treatment response, including: The metabolic parameters and their changes are processed using standard fractions to obtain standardized metabolic parameters. The standardized metabolic parameters were screened using intragroup correlation coefficients to obtain a set of standardized metabolic parameters related to treatment response.

4. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The stability screening of the radiomics features using intragroup correlation coefficients yields key features related to treatment response, including: Acquire two sets of image data of the target user at two time points within a preset time interval; The radiomics features of the two sets of image data were extracted separately, and the intra-group correlation coefficient of each radiomics feature was calculated. Radiomic features with intragroup correlation coefficients greater than a preset threshold were identified as key features associated with treatment response.

5. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The baseline clinical factors include age, sex, maximum standardized uptake, metabolic tumor volume, total glycolytic damage, TNM stage, histological differentiation, T stage, N stage, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and body mass index. Regression analysis of these baseline clinical factors identifies independent prognostic factors associated with treatment response, including: Univariate regression analysis was performed on each baseline clinical factor to obtain the first baseline clinical factor associated with survival outcome; Multivariate regression analysis was performed on the first baseline clinical factors to obtain the second baseline clinical factors associated with survival outcomes. The residual variable test was performed on the second baseline clinical factor to determine the second baseline clinical factor that satisfies the proportional risk assumption as the independent prognostic factor.

6. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The step of fusing the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the response rate of the target user includes: The standardized metabolic parameter set and the key features are weighted and fused to obtain a radiomics score; The radiomics score and the independent prognostic factors are then re-fused to obtain the response rate of the target user.

7. The method for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma according to claim 1, characterized in that, The step of fusing the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the prognostic prediction result for the target user includes: The standardized metabolic parameter set and the key features are weighted and fused to obtain a radiomics score; The radiomics score and the independent prognostic factors are fused using proportional hazards, and the fused hazard score that satisfies the proportional hazards assumption is determined as the prognostic prediction result for the target user.

8. A device for adjuvant chemoimmunotherapy response to oral squamous cell carcinoma, characterized in that, include: The data acquisition module is used to acquire the target user's imaging data and baseline clinical factors, including metabolic parameters and radiomics features of the tumor region. The standardization processing module is used to perform clinical parameter standardization processing on the metabolic parameters to obtain a set of standardized metabolic parameters related to treatment response; The stability screening module is used to screen the radiomics features for stability using intragroup correlation coefficients to obtain key features related to treatment response. The regression analysis module is used to perform regression analysis on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors; The treatment response module is used to perform feature fusion of the standardized metabolic parameter set, the key features, and the independent prognostic factors to obtain the response rate and prognostic prediction results of the target user.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the following steps: Acquire imaging data and baseline clinical factors of the target user, wherein the imaging data includes metabolic parameters and radiomics features of the tumor region; The metabolic parameters were standardized to clinical parameters to obtain a set of standardized metabolic parameters related to treatment response; Stability screening of the radiomics features was performed using intragroup correlation coefficients to obtain key features related to treatment response. Regression analysis was performed on the baseline clinical factors to identify independent prognostic factors related to treatment response among the baseline clinical factors; The standardized metabolic parameter set, the key features, and the independent prognostic factors are fused to obtain the response rate and prognostic prediction results for the target user.

10. A computer-readable storage medium, characterized in that, It contains a computer program that is loaded by a processor to perform the steps in the adjuvant chemoimmunotherapy response method for oral squamous cell carcinoma as described in any one of claims 1 to 7.

Images