Esophageal cancer multi-mode treatment effect comprehensive evaluation system

Abstract

The invention relates to the technical field of medical information, in particular to an esophageal cancer multi-mode treatment effect comprehensive evaluation system which comprises a patient data acquisition module, a drug resistance dynamic monitoring and adjusting module, an immune system response dynamic analysis module and a treatment effect evaluation module. The patient data acquisition module is used for acquiring multi-modal health data of a patient; the drug resistance dynamic monitoring and adjusting module analyzes drug resistance signs appearing in the treatment process; the immune system response dynamic analysis module comprehensively evaluates the state change of the immune system; and the treatment effect evaluation module evaluates the treatment effect of the patient in real time based on the drug resistance and immune system response analysis result, and provides treatment adjustment suggestions for the doctor according to the evaluation result. According to the invention, the accuracy of treatment is improved, the risk of treatment failure is reduced, and decision support based on data is provided for doctors.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to a comprehensive evaluation system for multimodal treatment effects of esophageal cancer. Background Art

[0002] With the continuous development of medical technology, precision medicine has played an increasingly important role in cancer treatment. As a common malignant tumor of the digestive system, the treatment effect of esophageal cancer is often affected by multiple factors, such as the patient's gene mutation, development of drug resistance, and immune system response. The comprehensive evaluation technology of multimodal data provides strong support for personalized treatment and can formulate the best treatment plan according to the patient's specific situation. However, real-time monitoring of the patient's treatment response and dynamic adjustment of the treatment plan is still a difficult problem in clinical practice.

[0003] Existing esophageal cancer treatment evaluation systems usually rely on only a single type of data, such as genetic testing data or imaging data, and are unable to effectively integrate multimodal health data for comprehensive evaluation. In addition, most evaluation systems lack real-time monitoring and feedback mechanisms during the treatment process, making it difficult to promptly identify changes in patients' drug resistance and immune system status. This results in doctors being unable to adjust treatment plans in a timely manner during the treatment process, thereby increasing the risk of treatment failure and lacking support for personalized treatment.

[0004] In response to the above problems, the present invention proposes a comprehensive evaluation system for the treatment effect of esophageal cancer based on multimodal data, which can not only improve the treatment effect, but also reduce the risk of side effects and treatment failure, and provide scientific decision-making support for doctors. Summary of the invention

[0005] The present invention provides a comprehensive evaluation system for multimodal treatment effects of esophageal cancer.

[0006] A multimodal comprehensive evaluation system for the treatment effect of esophageal cancer includes a patient data collection module, a drug resistance dynamic monitoring and adjustment module, an immune system response dynamic analysis module, and a treatment effect evaluation module, wherein;

[0007] The patient data collection module collects multimodal health data of the patient, including imaging data, genetic testing data, blood biochemical indicators, and immune system related data;

[0008] The drug resistance dynamic monitoring and adjustment module combines genetic test data and blood biochemical indicators to monitor the development trend of drug resistance of patients to therapeutic drugs in real time, analyze the signs of drug resistance that appear during the treatment process, and dynamically adjust the treatment plan according to the results of drug resistance analysis to provide treatment suggestions for doctors, including:

[0009] Drug resistance prediction: Combined with the patient's genetic test data, the Bayesian network algorithm is used to analyze the patient's gene mutation information to identify key gene mutation sites related to drug tolerance;

[0010] Real-time drug resistance monitoring: Based on the identified key gene mutation sites and combined with the patient's blood biochemical indicators, the development trend of drug resistance in patients during treatment is monitored in real time, changes in drug resistance signs are dynamically analyzed, and the effect of the treatment plan is judged by identifying the differences in the tolerance response of different drugs to patients;

[0011] Treatment adjustment: By analyzing the severity of drug resistance signs, the patient's treatment plan can be dynamically adjusted, including changing drugs, adjusting dosages, or implementing combination treatment plans;

[0012] The immune system response dynamic analysis module collects the patient's immune indicators in real time, comprehensively evaluates the changes in the state of the immune system, and dynamically monitors the response of the patient's immune system;

[0013] The treatment effect evaluation module evaluates the patient's treatment effect in real time based on the results of drug resistance and immune system response analysis, generates a treatment effect score, and provides doctors with treatment adjustment suggestions based on the evaluation results.

[0014] Optionally, the patient data collection module includes:

[0015] Image data acquisition: collect the patient's image data through imaging equipment (such as CT, MRI), and digitize and store the image data;

[0016] Genetic testing data collection: Genetic testing is performed on patients through genetic testing equipment to collect gene mutation data and expression information, and the genetic testing data is imported into the evaluation system through the data transmission interface;

[0017] Blood biochemical index data collection: blood biochemical indexes of patients, including enzymes, proteins and ions, are collected through blood testing equipment;

[0018] Immune system related data collection: Real-time monitoring of the patient's immune system related data through immune detection devices, including changes in immune cell levels and immune marker concentrations.

[0019] Optionally, the Bayesian network algorithm includes:

[0020] Constructing a Bayesian network structure related to gene mutation and drug resistance: Constructing a Bayesian network, where nodes represent the key gene mutation sites of patients and edges represent the dependencies between gene mutations, expressed as:

[0021]

[0022] Among them, X is all the gene mutation information of the patient, P(x i |Pa(x i )) represents gene x i The probability of x1, x2, ..., x under the condition of its parent node (i.e., the associated gene), n is the gene mutation site, n is the number of gene mutation sites;

[0023] Probability update based on gene mutation data: Use the patient's gene test data to update the Bayesian network probability and calculate the conditional probability distribution of each gene mutation node, expressed as:

[0024]

[0025] Among them, P(x i |Data) is the gene mutation x under the observed data conditions. i The posterior probability, P(Data|x i ) is a given x i The data likelihood, P(x i ) is the prior probability, P(Data) is the marginal probability of the data;

[0026] Drug resistance prediction: Based on the updated Bayesian network, the joint probability between each key gene mutation and drug resistance is calculated, and the gene mutation sites related to drug resistance are identified, which are expressed as:

[0027]

[0028] Among them, P(Y|X1,X2,...,X n ) means that given X1,X2,...,X n The conditional probability of event Y occurring when , Y is the patient's resistance to the drug (such as whether drug resistance occurs), X1, X2, ..., X n is a collection of multiple gene mutation sites of the patient, P(X1,X2,...,X n |Y) is the conditional probability of gene mutation given drug resistance, P(Y) is the prior probability of drug resistance, P(X1,X2,...,X n ) is the gene mutation X1,X2,...,X n The marginal probability of

[0029] Dynamically adjust the Bayesian network structure: Update the structure in real time according to the newly obtained genetic test data and biochemical indicators, and perform structural learning by minimizing the network's log-likelihood loss function, which is expressed as:

[0030]

[0031] in, is the loss function of the network structure, n is the total number of gene mutation nodes, x i is the i-th node in the Bayesian network.

[0032] Optionally, the real-time drug resistance monitoring includes:

[0033] Drug resistance probability calculation: Based on the identified key gene mutation sites X1, X2, ..., X n , combined with the patient's blood biochemical indicators B1, B2, ..., B m , calculate the conditional probability of each gene mutation being associated with drug resistance, expressed as:

[0034]

[0035] Among them, P(Y|X1,X2,...,X n ,B1,B2,...,B m ) represents the conditional probability of the patient's drug resistance status Y given the gene mutation and blood biochemical indicators, P(X1,X2,...,X n ,X n ,B1,B2,...,B m |Y) represents the joint probability of the patient's gene mutation and blood biochemical indicators under the condition of drug resistance state Y, P(Y) is the prior probability of drug resistance, P(X1,X2,...,X n ,B1,B2,...,B m ) is the marginal probability of gene mutation and blood biochemical indexes;

[0036] Real-time monitoring and dynamic analysis of changes in drug resistance signs: Real-time monitoring and dynamic analysis of changes in drug resistance signs, by identifying the dynamic change trends of gene mutation sites and biochemical indicators, assessing the probability of patients developing drug resistance during treatment, expressed as:

[0037] P(Y t+1 |X t ,B t )=αP(Y t |X t ,B t )+(1-α)P(X t ,B t );

[0038] Among them, P(Y t+1 |X t ,B t ) is at time t+1, combined with gene mutation X t and blood biochemical indexes Bt The predicted probability of drug resistance in patients, α is the adjustment coefficient;

[0039] Analysis of differences in drug tolerance responses and evaluation of treatment effects: By identifying the differences in patient tolerance responses to different drugs, the effectiveness of the current treatment regimen can be determined, expressed as:

[0040]

[0041] Among them, P(R|X,B,D) is the conditional probability of the patient's tolerance response R to drug D based on gene mutation data X and biochemical index B given drug D. X is the patient's gene mutation data, B is the patient's blood biochemical index, and D is the drug being evaluated. P(X,B|R,D) is the joint probability of the patient's gene mutation data X and biochemical index B given the patient's tolerance response R to drug D. P(R|D) is the prior probability that the patient will develop a tolerance response under drug D. P(X,B|D) is the marginal probability of the patient's gene mutation and biochemical indexes given drug D.

[0042] Optionally, the treatment regimen adjustment includes:

[0043] Evaluation of the severity of drug resistance signs: Based on the patient's gene mutation sites and blood biochemical indicators, the severity of drug resistance signs is calculated and expressed as:

[0044]

[0045] Among them, S represents the severity of drug resistance signs, w i and v j are the weights of gene mutation and biochemical index, P(x i |Pa(x i )) is the gene mutation site x i The probability under the condition of the parent node, P(b j |Y) is the biochemical index b j The probability under the condition of drug resistance Y, n is the number of gene mutation sites of the patient, and m is the number of blood biochemical indicators of the patient;

[0046] Dynamic adjustment of treatment plan: According to the results of drug resistance severity S, if S exceeds the preset threshold S threshold , adjust the patient's treatment plan in real time, including changing medications, adjusting the dosage of current medications, or implementing combination treatment plans.

[0047] Optionally, the immune system response dynamic analysis module includes:

[0048] Immune index collection: Real-time collection of patients' immune indicators, including the number of immune cells (such as T cells, B cells, etc.), immune marker levels (such as interleukins, tumor necrosis factors, etc.) and inflammatory indicators;

[0049] Immune system status assessment: The collected immune indicators are comprehensively evaluated through the immune system status assessment model to determine the patient's current immune system status, including the normal, overactive or immunosuppressed state of the patient's immune system;

[0050] Dynamic monitoring of immune response: By monitoring the changes in the immune system's response during treatment, the patient's immune response is dynamically tracked. By comparing with the baseline immune status, the immune system's adaptation to treatment is monitored. Combined with the dynamic changes in immune cells and markers, it is possible to predict whether the immune system is overreacting or having immune disorders.

[0051] Optionally, the immune index collection includes:

[0052] Immune cell count: Real-time detection of the patient's immune cell levels, including changes in the number of T cells and B cells, and continuous tracking of immune cell changes through blood analysis equipment or wearable devices;

[0053] Immune marker levels: Biochemical detection equipment is used to collect the patient's immune marker levels in real time, including interleukin and tumor necrosis factor;

[0054] Inflammatory markers: Monitor the patient's inflammatory response, including C-reactive protein (CRP), erythrocyte sedimentation rate, and levels of inflammatory markers related to the immune system.

[0055] Optionally, the immune system status assessment model adopts a deep belief network (DBN) model, and the deep belief network (DBN) model includes:

[0056] Input layer definition: the collected immune indicators are used as the input data of the model;

[0057] Restricted Boltzmann Machine (RBM) stacking: The deep belief network is composed of multiple restricted Boltzmann machines (RBMs) stacked together. Each layer of RBM consists of a visible layer and a hidden layer, and parameter updates are completed through unsupervised learning training.

[0058] Unsupervised layer-by-layer pre-training: The training of the deep belief network (DBN) model is unsupervised layer-by-layer pre-training. Each restricted Boltzmann machine (RBM) is trained in an unsupervised manner to learn the distribution of input data and extract the features of input data layer by layer. The output of each hidden layer is used as the input of the next layer, and is passed layer by layer until the last layer is trained.

[0059] Stacking multiple layers of RBM to build a deep network: By stacking multiple layers of RBM, the deep belief network (DBN) model gradually learns the multi-level representation of data;

[0060] Supervised fine-tuning: After unsupervised pre-training, the deep belief network (DBN) model is fine-tuned through supervised learning, using the labeled data of the immune system for the final classification task;

[0061] Output layer and classification: The output layer of the deep belief network (DBN) model classifies the state of the patient's immune system based on the characteristics of the immune indicators. (normal, overactive, or immunosuppressed).

[0062] Optionally, the dynamic monitoring of immune response includes:

[0063] Immune system response monitoring: real-time monitoring of the dynamic changes in the number of patients' immune cells (such as T cells, B cells, etc.) and immune marker levels (such as interleukins, tumor necrosis factor, etc.), and comparison with the patient's baseline immune status before treatment, and calculation of the difference between the immune system response and the baseline status;

[0064] Immune response trend analysis: Based on the difference between the immune system response and the baseline, the dynamic change trend of immune cells and markers is calculated, and the future change trend of the immune system is predicted using a time series analysis model;

[0065] Immune disorders and overreaction warning: Based on the changing trend of the immune system and the current treatment effect, it is determined whether the immune system is overreacting or immune disorder. When the immune system response changes exceed the set threshold I threshold When the patient is ill, an early warning is issued, indicating that the treatment plan needs to be adjusted.

[0066] Optionally, the treatment effect evaluation module includes:

[0067] Generate treatment effect score: Calculate the effect score S of the patient's current treatment regimen by combining the development trend of drug resistance and the adaptation of the immune system treatment ;

[0068] Provide treatment adjustment suggestions: Based on the generated treatment effect score S treatment , compared with the preset threshold θ threshold , to judge the effectiveness of the current treatment plan, if the score S treatment Below the set threshold θ threshold , then provide doctors with treatment adjustment suggestions, including suggestions for adjusting the type of medication, dosage or combination treatment plan.

[0069] Beneficial effects of the present invention:

[0070] The present invention collects and analyzes patients' health data in real time, including imaging data, genetic testing data, blood biochemical indicators and immune system-related data, and dynamically monitors drug resistance and analyzes immune system responses. The system can comprehensively track patients' treatment responses and provide dynamic personalized treatment plans. It can effectively improve the accuracy of treatment, reduce the risk of treatment failure, and provide doctors with data-based decision support.

[0071] The present invention, through the Bayesian network algorithm and deep belief network model, can extract key features from gene mutation and immune system data, accurately predict the patient's drug resistance development trend and immune system status changes. The system can monitor the patient's response to treatment in real time, dynamically adjust the treatment plan, ensure the effectiveness and safety of drugs and immunotherapy, thereby optimizing personalized treatment strategies and improving the overall treatment effect.

[0072] The present invention combines the analysis results of drug resistance and immune system response through a treatment effect evaluation module to generate a treatment effect score and provide real-time adjustment suggestions to doctors. This module ensures continuous optimization of treatment, reduces the risk of side effects, improves the success rate of treatment, and enhances the personalization and safety of treatment by dynamically evaluating the comprehensive treatment effects of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0074] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;

[0075] Figure 2 Schematic diagram of a drug resistance dynamic monitoring and adjustment module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0077] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0078] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0079] like Figure 1-Figure 2 As shown, a multimodal treatment effect comprehensive evaluation system for esophageal cancer includes a patient data collection module, a drug resistance dynamic monitoring and adjustment module, an immune system response dynamic analysis module and a treatment effect evaluation module, wherein;

[0080] The patient data collection module collects the patient's multimodal health data, including imaging data, genetic testing data, blood biochemical indicators, and immune system-related data;

[0081] The drug resistance dynamic monitoring and adjustment module combines genetic test data and blood biochemical indicators to monitor the development trend of drug resistance of patients to therapeutic drugs in real time, analyze the signs of drug resistance that appear during treatment, and dynamically adjust the treatment plan according to the results of drug resistance analysis to provide doctors with treatment suggestions, including:

[0082] Drug resistance prediction: Combined with the patient's genetic test data, the Bayesian network algorithm is used to analyze the patient's gene mutation information to identify key gene mutation sites related to drug tolerance;

[0083] Real-time drug resistance monitoring: Based on the identified key gene mutation sites and combined with the patient's blood biochemical indicators, the development trend of drug resistance in patients during treatment is monitored in real time, changes in drug resistance signs are dynamically analyzed, and the effect of the treatment plan is judged by identifying the differences in the tolerance response of different drugs to patients;

[0084] Treatment adjustment: By analyzing the severity of drug resistance signs, the patient's treatment plan can be dynamically adjusted, including changing drugs, adjusting dosages, or implementing combination treatment plans;

[0085] The immune system response dynamic analysis module collects the patient's immune indicators in real time, comprehensively evaluates the changes in the state of the immune system, and dynamically monitors the response of the patient's immune system;

[0086] The treatment effect evaluation module evaluates the treatment effect of patients in real time based on the results of drug resistance and immune system response analysis, generates a treatment effect score, and provides doctors with treatment adjustment suggestions based on the evaluation results to ensure the continuous optimization of the treatment plan;

[0087] Through the above content, it is possible to track the drug resistance and immune response that occurs during the patient's treatment in real time, dynamically adjust the treatment plan, ensure the accuracy and continuous optimization of the treatment, thereby improving the treatment effect, reducing the risk of treatment failure, and providing doctors with comprehensive decision-making support.

[0088] The patient data collection module includes:

[0089] Image data acquisition: collect the patient's image data through imaging equipment (such as CT, MRI), and digitize and store the image data;

[0090] Genetic testing data collection: Genetic testing is performed on patients through genetic testing equipment to collect gene mutation data and expression information, and the genetic testing data is imported into the evaluation system through the data transmission interface;

[0091] Blood biochemical index data collection: blood biochemical indexes of patients, including enzymes, proteins and ions, are collected through blood testing equipment;

[0092] Immune system related data collection: Real-time monitoring of the patient's immune system related data, including immune cell levels and immune marker concentration changes, through immune detection devices to ensure data continuity and integrity;

[0093] The above content can provide complete and accurate basic data for subsequent drug resistance monitoring, immune system analysis and treatment effect evaluation, ensuring that the system can comprehensively and dynamically track and optimize the patient's treatment response, provide doctors with scientific decision-making basis, and improve personalized treatment effects.

[0094] Bayesian network algorithms include:

[0095] Constructing a Bayesian network structure related to gene mutation and drug resistance: Constructing a Bayesian network, where nodes represent the key gene mutation sites of patients and edges represent the dependencies between gene mutations, expressed as:

[0096]

[0097] Among them, X is all the gene mutation information of the patient, P(x i |Pa(xi )) represents gene x i The probability of x1, x2, ..., x under the condition of its parent node (i.e., the associated gene), n is the gene mutation site, n is the number of gene mutation sites;

[0098] Probability update based on gene mutation data: Use the patient's gene test data to update the Bayesian network probability and calculate the conditional probability distribution of each gene mutation node, expressed as:

[0099]

[0100] Among them, P(x i |Data) is the gene mutation x under the observed data conditions. i The posterior probability, P(Data|x i ) is a given x i The data likelihood, P(x i ) is the prior probability, P(Data) is the marginal probability of the data;

[0101] Drug resistance prediction: Based on the updated Bayesian network, the joint probability between each key gene mutation and drug resistance is calculated, and the gene mutation sites related to drug resistance are identified, which are expressed as:

[0102]

[0103] Among them, P(Y|X1,X2,...,X n ) means that given X1,X2,...,X n The conditional probability of event Y occurring when , Y is the patient's resistance to the drug (such as whether drug resistance occurs), X1, X2, ..., X n is a collection of multiple gene mutation sites of the patient, P(X1,X2,...,X n |Y) is the conditional probability of gene mutation given drug resistance, P(Y) is the prior probability of drug resistance, P(X1,X2,...,X n ) is the gene mutation X1,X2,...,X n The marginal probability of represents the joint probability of gene mutation in the absence of drug resistance state Y;

[0104] Dynamically adjust the Bayesian network structure: Update the structure in real time according to the newly obtained genetic test data and biochemical indicators, and perform structural learning by minimizing the network's log-likelihood loss function, which is expressed as:

[0105]

[0106] in, is the loss function of the network structure, which represents the log-likelihood sum of each gene mutation node given its parent node, n is the total number of gene mutation nodes, x i is the i-th node in the Bayesian network, indicating the i-th gene mutation site;

[0107] Through the above content, it is possible to combine the patient's gene mutation information and blood biochemical indicators to achieve real-time prediction and monitoring of the development of drug resistance, which not only improves the accuracy and timeliness of drug resistance analysis, but also enables the system to accurately identify key gene mutation sites related to drug tolerance, providing doctors with a reliable basis for adjusting treatment plans, enhancing the adaptability and flexibility of the model, and helping to optimize personalized treatment plans for patients and improve overall treatment effects.

[0108] Real-time resistance monitoring includes:

[0109] Drug resistance probability calculation: Based on the identified key gene mutation sites X1, X2, ..., X n , combined with the patient's blood biochemical indicators B1, B2, ..., B m , calculate the conditional probability of each gene mutation being associated with drug resistance, expressed as:

[0110]

[0111] Among them, P(Y|X1,X2,...,X n ,B1,B2,...,B m ) represents the conditional probability of the patient's drug resistance status Y given the gene mutation and blood biochemical indicators, P(X1,X2,...,X n ,X n ,B1,B2,...,B m |Y) represents the joint probability of the patient's gene mutation and blood biochemical indicators in the case of drug resistance state Y, P(Y) is the prior probability of drug resistance, which represents the initial estimate of the patient's drug resistance when there is no data, P(X1,X2,...,X n ,B1,B2,...,B m ) is the marginal probability of gene mutation and blood biochemical index, representing their joint distribution;

[0112] Real-time monitoring and dynamic analysis of changes in drug resistance signs: Real-time monitoring and dynamic analysis of changes in drug resistance signs, by identifying the dynamic change trends of gene mutation sites and biochemical indicators, assessing the probability of patients developing drug resistance during treatment, expressed as:

[0113] P(Y t+1 |X t ,Bt )=αP(Y t |X t ,B t )+(1-α)P(X t ,B t );

[0114] Among them, P(Y t+1 |X t ,B t ) is at time t+1, combined with gene mutation X t and blood biochemical indexes B t The predicted probability of drug resistance in patients is α, which is the adjustment coefficient, weighing the importance of past and current data;

[0115] Analysis of differences in drug tolerance responses and evaluation of treatment effects: By identifying the differences in patient tolerance responses to different drugs, the effectiveness of the current treatment regimen can be determined, expressed as:

[0116]

[0117] Among them, P(R|X,B,D) is the conditional probability of the patient's tolerance response R to drug D based on gene mutation data X and biochemical index B given drug D. X is the patient's gene mutation data, B is the patient's blood biochemical index, and D is the drug being evaluated. P(X,B|R,D) is the joint probability of the patient's gene mutation data X and biochemical index B given the patient's tolerance response R to drug D. P(R|D) is the prior probability that the patient will develop a tolerance response under drug D. P(X,B|D) is the marginal probability of the patient's gene mutation and biochemical indexes given drug D.

[0118] Through the above content, it is possible to dynamically monitor the changing trend of drug resistance and timely analyze the differences in the responses of different drugs to patient tolerance. It can provide accurate drug resistance predictions in real time, help doctors identify drug resistance risks in advance during treatment, optimize treatment plans, avoid the risk of treatment failure, and improve the effectiveness and safety of personalized treatment.

[0119] Treatment adjustments include:

[0120] Evaluation of the severity of drug resistance signs: Based on the patient's gene mutation sites and blood biochemical indicators, the severity of drug resistance signs is calculated and expressed as:

[0121]

[0122] Among them, S represents the severity of drug resistance signs, w i and v j are the weights of gene mutation and biochemical index, P(x i|Pa(x i )) is the gene mutation site x i The probability under the condition of the parent node, P(b j |Y) is the biochemical index b j The probability under the condition of drug resistance Y, n is the number of gene mutation sites of the patient, and m is the number of blood biochemical indicators of the patient;

[0123] Dynamic adjustment of treatment plan: According to the results of drug resistance severity S, if S exceeds the preset threshold S threshold , adjust the patient's treatment plan in real time, including changing medications, adjusting the dosage of current medications, or implementing combination treatment plans;

[0124] The weights of gene mutations and biochemical indicators w i and v j The setting basis and value range include:

[0125] Gene mutation weight w i The setting of is based on the correlation between each gene mutation site and drug resistance. Based on historical data and medical research, gene mutation sites that are strongly associated with drug tolerance will receive higher weights, while gene mutation sites that have less impact on drug resistance will be given lower weights. The value range is [0,1];

[0126] Weight of biochemical index v j The basis for setting is the correlation between each biochemical indicator and the development trend of drug resistance. Some biochemical indicators (such as certain enzymes and protein levels in the blood) are directly related to drug metabolism and tolerance. These indicators will be given higher weights. In contrast, biochemical indicators with lower correlation are given smaller weights. The value range is [0,1].

[0127] Threshold S threshold Based on historical data and statistical analysis, it includes:

[0128] Collect historical data: Collect multimodal data of a large number of patients, including gene mutation information, biochemical indicators, drug resistance development and corresponding treatment effects;

[0129] Calculate the distribution of resistance signs: For each patient, calculate the severity S of resistance signs;

[0130] Determine a reasonable threshold: According to the S value distribution in historical data, select the appropriate threshold through statistical methods, and use the percentile method to select a percentile in the distribution as S. threshold , indicating that when S exceeds this value, it means that the signs of drug resistance have reached a serious level and the treatment plan needs to be adjusted, which is expressed as:

[0131] Sthreshold =P p (S);

[0132] Among them, P p (S) represents the pth percentile of the distribution of signs of drug resistance, and the 75% percentile was chosen to ensure that only in cases of more severe drug resistance would an adjustment in the treatment regimen be triggered;

[0133] Through the above content, drugs, dosages or combination treatment plans can be dynamically adjusted to ensure that the patient's treatment strategy can be optimized in a timely manner, thereby improving the success rate of treatment and reducing the risk of treatment failure due to drug resistance. It provides doctors with data-driven decision support and improves the accuracy and reliability of treatment effects.

[0134] The immune system response dynamic analysis module includes:

[0135] Immune index collection: Real-time collection of patients' immune indicators, including the number of immune cells (such as T cells, B cells, etc.), immune marker levels (such as interleukins, tumor necrosis factors, etc.) and inflammatory indicators;

[0136] Immune system status assessment: The collected immune indicators are comprehensively evaluated through the immune system status assessment model to determine the patient's current immune system status, including the normal, overactive or immunosuppressed state of the patient's immune system;

[0137] Dynamic monitoring of immune response: By monitoring the changes in the immune system's response during treatment, the patient's immune response is tracked dynamically. By comparing with the baseline immune status, the immune system's adaptation to treatment is monitored. Combined with the dynamic changes in immune cells and markers, it is possible to predict whether the immune system is overreacting or having immune disorders.

[0138] Through the above content, the comprehensive and accurate tracking of the patient's immune system is ensured, abnormal reactions of the immune system can be identified in advance, treatment plans can be adjusted dynamically, and the risk of immune imbalance during treatment can be reduced. Through data-driven dynamic feedback, this module provides more reliable decision support for personalized treatment and improves the effectiveness and safety of immunotherapy.

[0139] Immune index collection includes:

[0140] Immune cell count: Real-time detection of the patient's immune cell levels, including changes in the number of T cells and B cells, and continuous tracking of immune cell changes through blood analysis equipment or wearable devices;

[0141] Immune marker levels: Biochemical detection equipment is used to collect the patient's immune marker levels in real time, including interleukin and tumor necrosis factor;

[0142] Inflammatory indicators: Monitor the patient's inflammatory response, including C-reactive protein (CRP), erythrocyte sedimentation rate and levels of inflammatory markers related to the immune system;

[0143] Through the above content, doctors can ensure that they can understand the patient's immune status in a timely manner, quickly identify immune system abnormalities, and provide real-time data support for personalized treatment adjustments, which will help improve treatment outcomes and reduce the risk of immune response imbalance.

[0144] The immune system status assessment model uses a deep belief network (DBN) model, which includes:

[0145] Input layer definition: the collected immune indicators are used as the input data of the model;

[0146] Restricted Boltzmann Machine (RBM) stacking: The deep belief network is composed of multiple restricted Boltzmann machines (RBMs) stacked together. Each layer of RBM consists of a visible layer and a hidden layer. The parameters are updated through unsupervised learning training, which is expressed as:

[0147] E(v,h)=-∑ i a i v i -∑ j z j h j -∑ i,j v i W ij h j ;

[0148] Among them, E(v,h) is the energy function of the restricted Boltzmann machine RBM, v i and h j Represent the nodes of the visible layer and the hidden layer respectively, W ij is the weight connecting the visible layer and the hidden layer, a i and z j are the biases of the visible layer and the hidden layer respectively;

[0149] Unsupervised layer-by-layer pre-training: The training of the deep belief network (DBN) model is unsupervised layer-by-layer pre-training. Each restricted Boltzmann machine (RBM) is trained in an unsupervised manner to learn the distribution of input data and extract the features of input data layer by layer. The output of each hidden layer is used as the input of the next layer, and is passed layer by layer until the last layer is trained. It is expressed as:

[0150] The activation probability from the visible layer to the hidden layer: P(h j =1|v)=σ(∑ i v i W ij +z j );

[0151] The reconstruction probability from hidden layer to visible layer: P(v i =1|h)=σ(∑ j h j W ij +a i );

[0152] Among them, σ is the sigmoid activation function, which is used to calculate the activation probability;

[0153] Stacking multiple layers of RBM to build a deep network: By stacking multiple layers of RBM, the deep belief network (DBN) model gradually learns the multi-level representation of data, which is expressed as:

[0154] h (k) =σ(W (k) h (k-1) +z (k) );

[0155] Among them, h (k) is the output of the kth hidden layer, W (k) is the weight matrix of the kth layer, z (k) is the bias of the kth layer;

[0156] Supervised fine-tuning: After completing unsupervised pre-training, the deep belief network (DBN) model is fine-tuned through supervised learning, using the labeled data of the immune system for the final classification task, expressed as:

[0157]

[0158] Among them, L(θ) is the loss function, which represents the gap between the model prediction and the true label, y i is the true label of the i-th sample, is the predicted value of the model, θ is the parameters of the model (including weights and biases), and p is the number of training samples;

[0159] Output layer and classification: The output layer of the deep belief network (DBN) model classifies the state of the patient's immune system based on the characteristics of the immune indicators. (normal, overactive, or immunosuppressed), expressed as:

[0160]

[0161] in, is the predicted immune system state, W out is the weight matrix of the output layer, z out is the bias of the output layer, h (l) is the output of the last hidden layer;

[0162] Through the above content, it is possible to process complex high-dimensional data, such as the number of immune cells, levels of immune markers and inflammatory indicators. Its deep structure and unsupervised learning capabilities enable it to capture the complex changes in the state of the immune system and achieve accurate classification and evaluation of the normal, overactive or immunosuppressive state of the immune system. DBN is highly robust to noise and incomplete data, ensuring the accuracy and reliability of the assessment of the immune system status and helping to optimize personalized treatment plans.

[0163] Dynamic monitoring of immune response includes:

[0164] Immune system response monitoring: Real-time monitoring of the dynamic changes in the number of patients' immune cells (such as T cells, B cells, etc.) and immune marker levels (such as interleukins, tumor necrosis factor, etc.), and comparison with the patient's baseline immune status before treatment, calculate the difference between the immune system response and the baseline status, expressed as:

[0165]

[0166] Among them, ΔI represents the change in immune system response, I t,i is the i-th immune index value at time t, I baseline,i is the baseline value of the ith immune index, and n represents the number of immune indicators;

[0167] Immune response trend analysis: Based on the difference between the immune system response and the baseline, the dynamic change trend of immune cells and markers is calculated, and the future change trend of the immune system is predicted using a time series analysis model, which is expressed as:

[0168] P(I t+1 |I t )=γI t +(1-γ)I baseline ;

[0169] Among them, P(I t+1 |I t ) is the change trend of the immune system status at the predicted time t+1, and γ is the adjustment coefficient, which is used to balance the influence of current data and baseline data;

[0170] Immune disorders and overreaction warning: Based on the changing trend of the immune system and the current treatment effect, it is determined whether the immune system is overreacting or immune disorder. When the immune system response changes exceed the set threshold I threshold When the patient is ill, an early warning is issued, indicating that the treatment plan needs to be adjusted;

[0171] Threshold I threshold Based on statistical and clinical data analysis settings, specifically including:

[0172] Collect historical clinical data: Collect a large number of patients’ immune system response data during treatment, including immune cell numbers, immune marker levels, and inflammatory responses;

[0173] Calculate the range of changes in immune indicators: For each patient, calculate the range of changes in their immune indicators during treatment, and calculate the difference ΔI of each immune indicator based on the baseline value (immune system status before treatment);

[0174] Analyze the distribution of response differences: Based on the immune system response differences ΔI in historical data, construct the distribution of immune system response changes and determine the cutoff point between normal response and overreaction;

[0175] Set Threshold I threshold :By analyzing the quantiles in the distribution, an appropriate quantile (90%) is selected as the threshold, indicating that when the immune response exceeds this value, it indicates that an overreaction or immune disorder may occur, which is expressed as:

[0176] I threshold =P p (ΔI);

[0177] Among them, P p (ΔI) represents the pth quantile of the change ΔI of the immune response, and the commonly used quantile is 90%;

[0178] Through the above content, early warning can be issued to prevent adverse immune reactions during treatment, thereby helping doctors to adjust treatment plans in a timely manner, improving the personalization and safety of treatment, ensuring the healthy balance of the immune system, and optimizing the overall treatment effect of patients.

[0179] The treatment effect evaluation module includes:

[0180] Generate treatment effect score: Calculate the effect score S of the patient's current treatment regimen by combining the development trend of drug resistance and the adaptation of the immune system treatment , expressed as:

[0181] S treatment =w1S drug_resistance +w2S immune_response ;

[0182] Among them, S drug_resistance is the treatment effect score based on drug resistance assessment, S immune_response It is a treatment effect score based on the immune system response assessment, w1 and w2 are the weight parameters of drug resistance and immune response, respectively, w1+w2=1;

[0183]

[0184] Among them, R currentis the patient's current level of drug resistance development, determined by real-time monitoring of the patient's signs of drug resistance. max is the maximum threshold for the development of resistance, representing a situation of complete resistance;

[0185]

[0186] Among them, ΔI represents the difference between the immune response and the baseline state, representing the dynamic changes of the patient's immune system. max is the maximum tolerable range of changes in the immune system’s response;

[0187] The setting basis and value range of the weight parameters w1 and w2 of drug resistance and immune response specifically include:

[0188] The setting of the drug resistance weight parameter w1 is based on the correlation between drug resistance and treatment success rate in clinical data, and its value range is [0,1];

[0189] The weight parameter w2 of the immune response is set based on the clinical observation of the impact of the immune response on the overall treatment effect of the patient, and the value range is [0,1];

[0190] Provide treatment adjustment suggestions: Based on the generated treatment effect score S treatment , compared with the preset threshold θ threshold , to judge the effectiveness of the current treatment plan, if the score S treatment Below the set threshold θ threshold , then provide treatment adjustment suggestions to doctors, including suggestions for adjusting the type of medicine, dosage or combination treatment plan;

[0191] Threshold θ threshold Based on historical data and statistical analysis, it includes:

[0192] Collect historical data: Collect treatment effect data on a large number of patients, including drug resistance during treatment, immune system response data, and treatment outcomes;

[0193] Calculate historical scores: Based on historical cases, use the treatment effect score formula to calculate each patient's S treatment ;

[0194] Analyze the distribution of treatment effect scores: The patient's treatment effect score S treatment Count the successes and failures, draw their distribution graphs, and use statistical methods to analyze the distribution of scores to identify which score ranges correspond to treatment success or failure;

[0195] Set the threshold θ threshold: Based on historical data, an appropriate quantile is selected as the threshold, and the 90% quantile is selected to ensure that this value can better distinguish patients with good treatment effects and patients who need to adjust their treatment plans, expressed as:

[0196] θ threshold =P p (S treatment );

[0197] Among them, P p (S treatment ) represents the pth quantile of the treatment effect score, with a value of 90%;

[0198] Through the above content, the patient's treatment effect can be evaluated in real time and accurate effect scores can be generated. A comprehensive and dynamic evaluation mechanism is provided to help doctors adjust treatment plans in a timely manner to ensure the effectiveness and safety of the treatment. It can accurately identify potential problems in the treatment process and optimize personalized treatment strategies, thereby improving the patient's treatment success rate, reducing the risk of side effects, and enhancing the overall efficacy.

[0199] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0200] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A comprehensive evaluation system for multimodal treatment effects of esophageal cancer, characterized in that: It includes patient data collection module, drug resistance dynamic monitoring and adjustment module, immune system response dynamic analysis module and treatment effect evaluation module, among which; The patient data collection module collects multimodal health data of the patient, including imaging data, genetic testing data, blood biochemical indicators, and immune system related data; The drug resistance dynamic monitoring and adjustment module combines genetic test data and blood biochemical indicators to monitor the development trend of drug resistance of patients to therapeutic drugs in real time, analyze the signs of drug resistance that appear during the treatment process, and dynamically adjust the treatment plan according to the results of drug resistance analysis to provide treatment suggestions for doctors, including: Drug resistance prediction: Combined with the patient's genetic test data, the Bayesian network algorithm is used to analyze the patient's gene mutation information to identify key gene mutation sites related to drug tolerance; Real-time drug resistance monitoring: Based on the identified key gene mutation sites and combined with the patient's blood biochemical indicators, real-time monitoring of the patient's drug resistance development trend during treatment, dynamic analysis of changes in drug resistance signs, and identification of differences in the patient's tolerance response to different drugs to determine the effectiveness of the treatment plan; Treatment adjustment: By analyzing the severity of drug resistance signs, the patient's treatment plan can be dynamically adjusted, including changing drugs, adjusting dosages, or implementing combination treatment plans; The immune system response dynamic analysis module collects the patient's immune indicators in real time, comprehensively evaluates the changes in the state of the immune system, and dynamically monitors the response of the patient's immune system; The treatment effect evaluation module evaluates the patient's treatment effect in real time based on the results of drug resistance and immune system response analysis, generates a treatment effect score, and provides doctors with treatment adjustment suggestions based on the evaluation results.

2. The multimodal treatment effect comprehensive evaluation system for esophageal cancer according to claim 1, characterized in that: The patient data acquisition module comprises: Image data acquisition: collect the patient's image data through imaging equipment, and digitize and store the image data; Genetic testing data collection: Genetic testing is performed on patients through genetic testing equipment to collect gene mutation data and expression information, and the genetic testing data is imported into the evaluation system through the data transmission interface; Blood biochemical index data collection: blood biochemical indexes of patients, including enzymes, proteins and ions, are collected through blood testing equipment; Immune system related data collection: Real-time monitoring of the patient's immune system related data through immune detection devices, including changes in immune cell levels and immune marker concentrations.

3. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 1, characterized in that: The Bayesian network algorithm includes: Constructing a Bayesian network structure related to gene mutation and drug resistance: Constructing a Bayesian network, where nodes represent the key gene mutation sites of patients and edges represent the dependencies between gene mutations, expressed as: Among them, X is all the gene mutation information of the patient, P(x i |Pa(x i )) represents gene x i The probability of x1, x2, ..., x under the condition of its parent node n is the gene mutation site, n is the number of gene mutation sites; Probability update based on gene mutation data: Use the patient's gene test data to update the Bayesian network probability and calculate the conditional probability distribution of each gene mutation node, expressed as: Among them, P(x i |Data) is the gene mutation x under the observed data conditions. i The posterior probability, P(Data|x i ) is a given x i The data likelihood, P(x i ) is the prior probability, P(Data) is the marginal probability of the data; Drug resistance prediction: Based on the updated Bayesian network, the joint probability between each key gene mutation and drug resistance is calculated, and the gene mutation sites related to drug resistance are identified, which are expressed as: Among them, P(Y|X1,X2,...,X n ) means that given X1,X2,...,X n The conditional probability of event Y occurring when , Y is the patient's resistance to the drug, X1, X2, ..., X n is a collection of multiple gene mutation sites of the patient, P(X1,X2,...,X n |Y) is the conditional probability of gene mutation given drug resistance, P(Y) is the prior probability of drug resistance, P(X1,X2,...,X n ) is the gene mutation X1,X2,...,X n The marginal probability of Dynamically adjust the Bayesian network structure: Update the structure in real time according to the newly obtained genetic test data and biochemical indicators, and perform structural learning by minimizing the network's log-likelihood loss function, which is expressed as: in, is the loss function of the network structure, n is the total number of gene mutation nodes, x i is the i-th node in the Bayesian network.

4. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 3, characterized in that: The real-time drug resistance monitoring includes: Drug resistance probability calculation: Based on the identified key gene mutation sites X1, X2, ..., X n , combined with the patient's blood biochemical indicators B1, B2, ..., B m , calculate the conditional probability of each gene mutation being associated with drug resistance, expressed as: Among them, P(Y|X1,X2,...,X n ,B1,B2,...,B m ) represents the conditional probability of the patient's drug resistance status Y given the gene mutation and blood biochemical indicators, P(X1,X2,...,X n ,X n ,B1,B2,...,B m |Y) represents the joint probability of the patient's gene mutation and blood biochemical indicators under the condition of drug resistance state Y, P(Y) is the prior probability of drug resistance, P(X1,X2,...,X n ,B1,B2,...,B m ) is the marginal probability of gene mutation and blood biochemical indexes; Real-time monitoring and dynamic analysis of changes in drug resistance signs: Real-time monitoring and dynamic analysis of changes in drug resistance signs, by identifying the dynamic change trends of gene mutation sites and biochemical indicators, assessing the probability of patients developing drug resistance during treatment, expressed as: P(Y t+1 |X t ,B t )=αP(Y t |X t ,B t )+(1-α)P(X t ,B t ); Among them, P(Y t+1 |X t ,B t ) is at time t+1, combined with gene mutation X t and blood biochemical indexes B t The predicted probability of drug resistance in patients, α is the adjustment coefficient; Analysis of drug tolerance response differences and evaluation of treatment effects: By identifying the differences in the tolerance responses of different drugs to patients, the effect of the current treatment plan can be determined, expressed as: Among them, P(R|X,B,D) is the conditional probability of the patient's tolerance response R to drug D based on gene mutation data X and biochemical index B given drug D. X is the patient's gene mutation data, B is the patient's blood biochemical index, and D is the drug being evaluated. P(X,B|R,D) is the joint probability of the patient's gene mutation data X and biochemical index B given the patient's tolerance response R to drug D. P(R|D) is the prior probability that the patient will develop a tolerance response under drug D. P(X,B|D) is the marginal probability of the patient's gene mutation and biochemical indexes given drug D.

5. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 4, characterized in that: The treatment regimen adjustments include: Evaluation of the severity of drug resistance signs: Based on the patient's gene mutation sites and blood biochemical indicators, the severity of drug resistance signs is calculated and expressed as: Among them, S represents the severity of drug resistance signs, w i and v j are the weights of gene mutation and biochemical index, P(x i |Pa(x i )) is the gene mutation site x i The probability under the parent node condition, P(b j |Y) is the biochemical index b j The probability under the condition of drug resistance Y, n is the number of gene mutation sites of the patient, and m is the number of blood biochemical indicators of the patient; Dynamic adjustment of treatment plan: According to the results of drug resistance severity S, if S exceeds the preset threshold S threshold , adjust the patient's treatment plan in real time, including changing medications, adjusting the dosage of current medications, or implementing combination treatment plans.

6. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 1, characterized in that: The immune system response dynamic analysis module includes: Immune index collection: Real-time collection of patients’ immune indicators, including immune cell count, immune marker levels, and inflammatory indicators; Immune system status assessment: The collected immune indicators are comprehensively evaluated through the immune system status assessment model to determine the patient's current immune system status, including the normal, overactive or immunosuppressed state of the patient's immune system; Dynamic monitoring of immune response: By monitoring the changes in the immune system's response during treatment, the patient's immune response is dynamically tracked. By comparing with the baseline immune status, the immune system's adaptation to treatment is monitored. Combined with the dynamic changes in immune cells and markers, it is possible to predict whether the immune system is overreacting or having immune disorders.

7. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 6, characterized in that: The immune index collection includes: Immune cell count: Real-time detection of the patient's immune cell levels, including changes in the number of T cells and B cells, and continuous tracking of immune cell changes through blood analysis equipment or wearable devices; Immune marker levels: Biochemical detection equipment is used to collect the patient's immune marker levels in real time, including interleukin and tumor necrosis factor; Inflammatory markers: Monitor the patient's inflammatory response, including C-reactive protein, erythrocyte sedimentation rate, and levels of inflammatory markers related to the immune system.

8. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 7, characterized in that: The immune system status assessment model adopts a deep belief network model, and the deep belief network model includes: Input layer definition: the collected immune indicators are used as input data of the model; Restricted Boltzmann Machine (RBM) stacking: The deep belief network is composed of multiple restricted Boltzmann machines (RBMs) stacked together. Each layer of RBM consists of a visible layer and a hidden layer, and parameter updates are completed through unsupervised learning training. Unsupervised layer-by-layer pre-training: The training of the deep belief network model is unsupervised layer-by-layer pre-training. Each restricted Boltzmann machine (RBM) is trained in an unsupervised manner to learn the distribution of input data and extract the features of input data layer by layer. The output of each hidden layer is used as the input of the next layer, and is passed layer by layer until the last layer is trained. Stacking multiple layers of RBM to build a deep network: By stacking multiple layers of RBM, the deep belief network model gradually learns the multi-level representation of data; Supervised fine-tuning: After unsupervised pre-training, the deep belief network model is fine-tuned through supervised learning, using labeled data from the immune system for the final classification task; Output layer and classification: The output layer of the deep belief network model classifies the state of the patient's immune system based on the characteristics of the immune indicators.

9. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 8, characterized in that: The dynamic monitoring of immune response includes: Immune system response monitoring: real-time monitoring of the dynamic changes in the number of immune cells and immune marker levels of patients, and comparison with the baseline immune status of patients before treatment, to calculate the difference between the immune system response and the baseline status; Immune response trend analysis: Based on the difference between the immune system response and the baseline, the dynamic change trend of immune cells and markers is calculated, and the future change trend of the immune system is predicted using a time series analysis model; Immune disorders and overreaction warning: Based on the changing trend of the immune system and the current treatment effect, it is determined whether the immune system is overreacting or immune disorder. When the immune system response changes exceed the set threshold I threshold When the patient is ill, an early warning is issued, indicating that the treatment plan needs to be adjusted.

10. The comprehensive evaluation system for multimodal treatment effect of esophageal cancer according to claim 1, characterized in that: The treatment effect evaluation module includes: Generate treatment effect score: Calculate the effect score S of the patient's current treatment regimen by combining the development trend of drug resistance and the adaptation of the immune system treatment ; Provide treatment adjustment suggestions: Based on the generated treatment effect score S treatment , compared with the preset threshold θ threshold , to judge the effectiveness of the current treatment plan, if the score S treatment Below the set threshold θ threshold , then provide doctors with treatment adjustment suggestions, including suggestions for adjusting the type of medication, dosage or combination treatment plan.

Images