Liver cancer treatment effect analysis system and method based on metabolic comprehensive analysis

Abstract

The present application relates to the technical field of physiological monitoring, and particularly relates to a hepatocellular carcinoma treatment effect analysis system and method based on metabolic comprehensive analysis, which comprises the following steps: obtaining the evolution of hepatocellular carcinoma, the composition of drugs and the composition of metabolites; performing drug absorption and conversion analysis based on the composition of drugs and the composition of metabolites and the marker condition; predicting future drug conversion components based on the drug absorption and conversion analysis results and the evolution of hepatocellular carcinoma; and performing drug treatment effect early warning based on the future drug conversion component prediction, correcting the first-pass effect through mass spectrometry imaging and enzyme activity, quantifying the inhibition effect of the tumor microenvironment on drug conversion, and dynamically adjusting the prediction by using the future tumor abnormal score ratio, thereby improving the accuracy of the timing of drug replacement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of physiological monitoring, and particularly relates to a hepatocellular carcinoma treatment effect analysis system and method based on metabolic comprehensive analysis. BACKGROUND

[0002] Hepatocellular carcinoma (HCC) is one of the most common and high-mortality malignant tumors worldwide, seriously threatening human health. In recent years, although certain progress has been made in the treatment of hepatocellular carcinoma, such as the application of surgical resection, chemotherapy, radiotherapy, targeted therapy, and immunotherapy, the overall treatment effect is still unsatisfactory, and the survival rate and quality of life of patients face many difficulties. Traditional treatment methods often lack sufficient consideration of tumor heterogeneity and dynamic changes. Hepatocellular carcinoma has complex biological characteristics, and tumors in different patients differ significantly in gene expression, metabolic characteristics, and microenvironment. This heterogeneity makes the same treatment regimen have completely different effects on different patients, leading to some patients failing to benefit from existing treatments, and even tumor progression and recurrence. Meanwhile, tumor cells continue to evolve during treatment, leading to drug resistance, which is one of the important reasons for the failure of hepatocellular carcinoma treatment. For example, when using the targeted drug sorafenib to treat hepatocellular carcinoma, some patients will develop sorafenib resistance due to CYP3A4 inactivation, making the drug unable to exert its full effect, and the tumor continues to grow and metastasize.

[0003] Drug metabolism is crucial for drug efficacy and safety. In the treatment of hepatocellular carcinoma, drugs need to undergo a series of complex metabolic processes after entering the body, including absorption, distribution, metabolism, and excretion. However, there are still many deficiencies in the study of drug metabolism mechanisms and patterns in hepatocellular carcinoma patients. Traditional pharmacokinetic studies are mainly based on population pharmacokinetic models, which often ignore individual differences and the influence of tumor microenvironment on drug metabolism. The tumor microenvironment is a complex ecosystem that includes tumor cells, immune cells, stromal cells, and various cytokines and metabolites. Some metabolites such as lactic acid can inhibit the transformation process of drugs and affect drug efficacy. However, these tumor microenvironment factors are rarely considered in pharmacokinetic studies, leading to inaccurate prediction of drug metabolism and efficacy in patients. In addition, existing drug metabolism research methods are mostly static and cannot monitor the dynamic changes of drug metabolism in real time. Drug metabolism in the body is a dynamic and time-varying process, and the metabolites and concentrations of drugs at different time points may change greatly. Traditional research methods can only sample and analyze at limited time points, and cannot comprehensively and timely reflect the dynamic process of drug metabolism, making it difficult to meet the needs of precise clinical treatment.

[0004] To solve these problems, the application designs a hepatocellular carcinoma treatment effect analysis system and method based on metabolic comprehensive analysis. SUMMARY

[0005] The application aims to provide a hepatocellular carcinoma treatment effect analysis system and method based on metabolic comprehensive analysis. The application integrates the evolution data of hepatocellular carcinoma, the composition of drugs and the dynamics of metabolites, combines deep learning to construct a multi-modal time series prediction model, realizes dynamic correlation analysis of drug metabolism pathways and tumor abnormality scores, corrects the first-pass effect through mass spectrometry imaging and enzyme activity, quantifies the inhibition effect of the tumor microenvironment on drug conversion, dynamically adjusts the prediction using future tumor abnormality score ratios, and improves the accuracy of drug replacement timing.

[0006] The application is implemented as follows:

[0007] In a first aspect, the application provides a hepatocellular carcinoma treatment effect analysis method based on metabolic comprehensive analysis, including the following steps:

[0008] S1, obtaining the evolution of hepatocellular carcinoma, the composition of drugs and the composition of metabolites;

[0009] S2, performing drug absorption and conversion analysis based on the labeling of the composition of drugs and the composition of metabolites;

[0010] S3, predicting future drug conversion components based on the drug absorption and conversion analysis results and the evolution of hepatocellular carcinoma;

[0011] S4, performing drug treatment effect early warning based on the future drug conversion component prediction.

[0012] As an implementation manner of the application, the evolution of hepatocellular carcinoma in step S1 includes images before and after drug application, the tumor evolution track is tracked through continuous imaging examination, the change of the tumor is obtained, and the metabolic characteristics of the tumor microenvironment are obtained, including lactic acid and glutamine levels. The composition of the drug includes the composition, concentration, half-life and action tissue of the drug used, and the composition of the metabolite includes the drug concentration in the metabolite and the component concentration after drug conversion in the corresponding metabolite.

[0013] As an implementation manner of the application, the drug absorption and conversion analysis in step S2 includes the following specific steps:

[0014] S21, obtain the composition of the drug and the composition of the metabolite, obtain the composition of the consumed drug, subtract the corresponding composition of the remaining metabolite from the composition of the consumed drug, obtain the distribution of various consumed drug compositions in each position of the liver, and obtain the conversion path of the consumed drug composition and the generated metabolite based on the constructed enzyme reaction-based metabolic path, the specific steps are: obtaining the concentration of historical enzymes, the survival environment of enzymes, the composition of the consumed drug, the composition of the metabolite, and the metabolic conversion path of the drug, constructing a deep learning neural network model with the concentration of enzymes, the survival environment of enzymes, the composition of the consumed drug, and the composition of the metabolites as inputs, and the metabolic conversion path of the drug as output, analyzing the conversion path of the drug to the metabolite based on the constructed deep learning neural network model, the conversion path includes the conversion of the drug in each tissue of the liver, and the actual consumption of the drug is quantitatively calculated by accurately determining the content of the drug prototype and the metabolite, combined with the drug residues in the excreta and serum, which can correct the influence of the liver first-pass effect, ensure the accuracy of the metabolic kinetics data, and provide a reliable basis for subsequent modeling, using mass spectrometry imaging technology, the concentration gradient distribution of the drug in different functional areas of the liver can be visualized, and the difference in penetration of the drug in tumor tissue and normal tissue is revealed. This helps to understand the targeting efficiency of the drug, and provides a basis for optimizing the drug administration scheme, combined with metabolic enzyme activity, pH, temperature and other microenvironment parameters, the LSTM neural network is used to dynamically predict the drug metabolism path, which improves the prediction accuracy;

[0015] S22, simultaneously obtain the tumor evolution of each tissue of the liver, and perform tumor abnormality judgment based on the size, mutation load and image change of the tumor evolution of each tissue;

[0016] The specific steps are: obtaining the size, mutation load and image change of the tumor evolution of each tissue, wherein the image change is the average value of the difference between the pixel value of the pixel point and the pixel value of the normal tissue, and the standardized value is obtained by dividing the corresponding standard value, and the corresponding tumor abnormality judgment result is obtained by weighted summation; wherein the weighting weight is obtained by historical data experiment, that is, the tumor of the corresponding tissue destroys the corresponding tissue, so that the corresponding tissue cannot absorb and convert the drug, and the damage degree of the tissue function is quantified by comprehensively analyzing the mutation, image and other data, reflecting the decrease of drug metabolism capacity, for example, tumor destroys cytochrome P450 (such as CYP3A4) to cause metabolic disorder;

[0017] S23, analyze the drug conversion and absorption of the corresponding conversion path through the consumption of the drug and the tumor abnormality judgment of the corresponding tissue;

[0018] Specific steps are: obtaining the tumor abnormality judgment result of each tissue on the conversion path of the drug conversion into metabolites, and the consumption amount of the corresponding drug;

[0019] The tumor abnormality judgment result on the conversion path is obtained by weighted summation based on the tumor abnormality judgment result of each tissue, and the weighted weight is distributed by the proportion of the drug conversion amount of each tissue in the total drug conversion amount;

[0020] The drug consumption coefficient is obtained by dividing the drug consumption amount by the corresponding drug consumption standard value, the corresponding drug conversion analysis result is obtained by dividing the drug consumption coefficient by the tumor abnormality judgment result, the inhibition effect of the tumor on the metabolic path is directly evaluated by the ratio of the drug consumption coefficient to the tumor score, the higher the proportion of abnormal tissues, the lower the drug conversion rate, and the weighted summation and standardization are general methods of multivariate data analysis.

[0021] As an implementation manner of the present application, the future drug conversion component prediction in step S3 specifically includes the following steps:

[0022] S31, obtaining the conversion analysis result of the corresponding drug and the tumor abnormality judgment result on the corresponding conversion path; integrating the drug conversion analysis result (such as metabolite concentration, enzyme activity) and the tumor abnormality score (such as imageomics features, mutation load) to construct a multi-modal time series data set, according to: drug metabolism kinetics research shows that the tumor microenvironment (such as hypoxic state) will significantly affect the drug conversion efficiency;

[0023] S32, predicting the tumor abnormality judgment result on the future conversion path based on the historical conversion analysis result of the corresponding drug and the tumor abnormality judgment result on the conversion path, wherein the prediction of the tumor abnormality judgment result on the future conversion path is performed by a deep learning neural network;

[0024] S33, obtaining the ratio of the predicted tumor abnormality judgment result on the future conversion path to the current tumor abnormality judgment result, obtaining the future drug conversion analysis prediction result by dividing the corresponding drug conversion analysis result by the corresponding ratio, and taking the ratio of the future tumor abnormality score to the current tumor abnormality score as a correction factor of the drug conversion efficiency; according to: the tumor load change is negatively correlated with the drug metabolism clearance rate; advantage: the prediction result is adjusted by the dynamic ratio to avoid misjudgment caused by the static threshold.

[0025] As an implementation manner of the present application, the drug treatment effect early warning based on the future drug conversion component prediction in step S4 includes the following specific steps:

[0026] The future drug conversion analysis prediction result is compared with the set drug conversion analysis threshold value, if the corresponding future drug conversion analysis prediction result is greater than or equal to the set drug conversion analysis threshold value, it indicates that the drug effect is appropriate, and no drug replacement warning is needed, if the corresponding future drug conversion analysis prediction result is less than the set drug conversion analysis threshold value, it indicates that the drug effect is not appropriate, and drug replacement warning is needed, the prediction result is compared with the drug conversion threshold value, triggering the warning, the threshold value is usually obtained based on the historical pharmacodynamics or clinical drug resistance threshold value, and the prediction of metabolic efficiency reduction can trigger drug replacement in advance.

[0027] In a second aspect, the present application provides a hepatocellular carcinoma treatment effect analysis system based on metabolic comprehensive analysis, comprising:

[0028] A data acquisition module acquires the evolution of hepatocellular carcinoma, the composition of the drug and the composition of the metabolite;

[0029] A drug absorption and conversion analysis module performs drug absorption and conversion analysis based on the composition of the drug and the composition of the metabolite;

[0030] A future drug conversion component prediction module predicts future drug conversion components based on the drug absorption and conversion analysis result and the evolution of the corresponding hepatocellular carcinoma;

[0031] A treatment effect warning module performs drug treatment effect warning based on the future drug conversion component prediction.

[0032] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes a hepatocellular carcinoma treatment effect analysis method based on metabolic comprehensive analysis by calling the computer program stored in the memory.

[0033] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0034] The present application integrates the evolution data of hepatocellular carcinoma, the drug composition and the metabolite dynamics, combines deep learning to construct a multi-modal time series prediction model, realizes dynamic correlation analysis of drug metabolism path and tumor abnormal score, corrects the first-pass effect through mass spectrometry imaging and enzyme activity, quantifies the inhibition effect of tumor microenvironment on drug conversion, dynamically adjusts the prediction by using the future tumor abnormal score ratio, and improves the accuracy of drug replacement timing. BRIEF DESCRIPTION OF DRAWINGS

[0035] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0036] Figure 1A schematic diagram of the overall process of the liver cancer treatment effect analysis method based on metabolic comprehensive analysis of the present application;

[0037] Figure 2 A schematic diagram of the analysis process of step S3 of the liver cancer treatment effect analysis method based on metabolic comprehensive analysis of the present application;

[0038] Figure 3 An analysis flowchart of drug transformation absorption analysis of the transformation path of the liver cancer treatment effect analysis method based on metabolic comprehensive analysis of the present application;

[0039] Figure 4 A structure diagram of the liver cancer treatment effect analysis system based on metabolic comprehensive analysis of the present application. DETAILED DESCRIPTION

[0040] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments and the embodiments can be combined with each other.

[0041] Embodiment 1

[0042] As shown in Figures 1 to 3 , the present embodiment provides a liver cancer treatment effect analysis method based on metabolic comprehensive analysis, which specifically includes the following steps:

[0043] S1, obtaining the evolution of liver cancer, the composition of the drug and the composition of the metabolite;

[0044] Exemplarily, the evolution of liver cancer in step S1 includes images before and after the drug is applied. The tumor evolution track is tracked through continuous imaging examination, and the change of the tumor is obtained, and the metabolic characteristics of the tumor microenvironment are obtained, including lactic acid and glutamine levels. The composition of the drug includes the composition, concentration, half-life and action tissue of the drug used, for example, the position of sorafenib acting on the liver and the concentration in the liver. The composition of the metabolite includes the concentration of the drug in the metabolite and the concentration of the components formed after the drug transformation in the corresponding metabolite, for example, the component generated by the reaction of the effective component in the drug acting on the tumor position and the cancer cells. This can be obtained from the isotope labeling method, and carbon-14 labeled drug can be used to track the metabolic path in the liver cancer tissue;

[0045] Specifically, patients diagnosed with hepatocellular carcinoma are selected from hospitals, and suitable research subjects are screened according to inclusion and exclusion criteria. The inclusion criteria can include a clear diagnosis of hepatocellular carcinoma, no specific treatment, etc.; the exclusion criteria can include the combination of other serious diseases, the inability to cooperate with the examination, etc., and the patients are divided into different groups according to the treatment plan or research purpose, such as the drug treatment group and the control group;

[0046] Prepare CT, MRI and other imaging examination equipment, and debug and calibrate to ensure image quality and accuracy; prepare reagents for detecting tumor microenvironment metabolic characteristics, such as kits for detecting lactic acid and glutamine levels; prepare carbon-14 labeled drugs to ensure the stability and accuracy of the label; prepare instruments for analyzing drug composition, concentration, metabolite composition, etc., such as high-performance liquid chromatographs (HPLC), mass spectrometers (MS), etc.;

[0047] Before drug application, CT or MRI examination is performed on all patients to obtain initial image information of the tumor, including tumor size, location, shape, etc.; during drug treatment, imaging examination is performed on patients at predetermined time intervals (such as every week, every two weeks, or every month) to track the evolution trajectory of the tumor and record changes in tumor size, shape, boundary, etc.

[0048] At each imaging examination, tumor tissue or blood samples are collected from patients for detection of metabolic characteristics of the tumor microenvironment; using corresponding kits or detection methods, the levels of metabolites such as lactic acid and glutamine in the sample are detected; the composition and concentration of the drug used are recorded, and blood or tissue samples are collected from patients regularly during treatment, and HPLC or MS methods are used to detect the concentration of the drug in the body; by collecting samples multiple times, the concentration of the drug in the body changes with time, and the half-life of the drug is calculated; combined with the imaging examination and drug concentration detection results, the tissue location of the drug action and the concentration in the corresponding tissue are determined, such as the location of sorafenib in the liver and the concentration in the liver;

[0049] The drug treatment group patients are given carbon-14 labeled drugs according to the predetermined dose and administration method; at different time points after drug administration, tumor tissue or blood samples are collected from patients, and HPLC-MS methods are used to analyze the concentration of drugs in metabolites and the concentration of components formed after drug conversion, to track the metabolic pathway of drugs in hepatocellular carcinoma tissue;

[0050] S2, based on the composition of the drug and the composition of the metabolites, the drug absorption and conversion analysis is carried out;

[0051] The specific steps are: S21, obtaining the composition of the drug and the composition of the metabolite, obtaining the consumed drug composition, subtracting the remaining corresponding component composition in the metabolite from the component composition of the consumed drug to obtain the consumed drug composition, obtaining the distribution of various consumed drug compositions in each position of the liver, obtaining the conversion path of the consumed drug composition and the generated metabolite based on the constructed enzyme reaction-based metabolic path, the specific steps are: obtaining the concentration of historical enzymes, the survival environment of enzymes (including temperature, humidity and pH, etc.), the component composition of the consumed drug, the component composition in the metabolite and the metabolic conversion path of the drug, constructing a deep learning neural network model with the concentration of enzymes, the survival environment of enzymes, the component composition of the consumed drug and the component composition in the metabolite as input and the metabolic conversion path of the drug as output, analyzing the conversion path of the drug to the metabolite based on the constructed deep learning neural network model, the conversion path includes the conversion of the drug in each tissue of the liver, the actual consumption of the drug is quantitatively calculated by accurately determining the content of the drug prototype and the metabolite, combining the drug residues in the excretion and serum, this method can correct the influence of liver first-pass effect, ensure the accuracy of metabolic kinetics data, provide a reliable basis for subsequent modeling, using mass spectrometry imaging technology, the concentration gradient distribution of the drug in different functional areas of the liver can be visualized, and the difference in penetration of the drug in tumor tissue and normal tissue is revealed. This helps to understand the targeting efficiency of the drug, and provides a basis for optimizing the drug delivery scheme (such as local delivery), combined with metabolic enzyme activity, pH, temperature and other microenvironment parameters, the LSTM neural network is used to dynamically predict the drug metabolism path, which improves the accuracy of prediction;

[0052] The specific content of the neural network is: first, the data collection stage: comprehensively collect enzyme-related data, including the concentration of enzymes, the survival environment of enzymes, such as temperature, humidity and pH value, etc. These factors have a significant impact on the activity of enzymes and the process of drug metabolism, at the same time, the component composition of the consumed drug and the specific component composition in the metabolite are recorded in detail, in addition, the metabolic conversion path of the drug is studied and recorded in depth, which provides a rich and accurate data basis for subsequent model training;

[0053] Then data division is carried out, and the collected historical data is divided according to the proportion of 85% and 15%, wherein 85% of the data is used as the weight, bias training set for model training; and the other 15% of the data is used as the weight, bias test set for performance evaluation of the trained model; then is the model training stage. The 85% weight, bias training set is input into the deep learning neural network model, and in the training process, the model continuously adjusts its parameters and structure to learn the rules and characteristics in the data. Through multiple iterations and optimization, the initial deep learning neural network model is obtained; finally is the model testing and screening, using the 15% weight, bias test set to test the initial deep learning neural network model, and in the testing process, the accuracy of the model in judging the drug metabolism and transformation path is focused on. By continuously comparing the performance of different models, the initial deep learning neural network model that can meet the maximum metabolism and transformation path judgment accuracy is selected as the final deep learning neural network model;

[0054] S22, simultaneously acquire the tumor evolution of each tissue of the liver, and perform tumor abnormality judgment based on the size, mutation load and image change of the tumor evolution of each tissue;

[0055] The specific steps are: acquiring the size, mutation load and image change of the tumor evolution of each tissue, wherein the image change is the average value of the difference between the pixel value of the pixel point and the pixel value of the normal tissue, and the standardized value is obtained after the average value is divided by the corresponding standard value; the corresponding tumor abnormality judgment result is obtained after weighted summation; wherein the weighting weight is obtained through historical data experiment, that is, the tumor of the corresponding tissue destroys the corresponding tissue, so that the corresponding tissue cannot absorb and transform the drug, and the damage degree of the tissue function is quantified by comprehensively analyzing the mutation, image and other data, reflecting the decrease of drug metabolism capacity, for example, the tumor destroys the cytochrome P450 (such as CYP3A4) of the liver cell, resulting in metabolic disorder;

[0056] S23, performing drug transformation and absorption analysis of the corresponding transformation path according to the consumption of the drug and the tumor abnormality judgment of the corresponding tissue;

[0057] The specific steps are: acquiring the tumor abnormality judgment result of each tissue on the transformation path of the drug to the metabolite, and the consumption amount of the corresponding drug;

[0058] The tumor abnormality judgment result on the transformation path is obtained by weighted summation based on the tumor abnormality judgment result of each tissue, and the weighting weight is distributed by the proportion of the drug transformation amount of each tissue in the total drug transformation amount;

[0059] The drug consumption coefficient is obtained by dividing the drug consumption amount by the corresponding drug consumption standard value, the corresponding drug conversion analysis result is obtained by dividing the drug consumption coefficient by the tumor abnormality judgment result, the inhibition effect of the tumor on the metabolic pathway is directly evaluated by the ratio of the drug consumption coefficient to the tumor score, the higher the proportion of abnormal tissues, the lower the drug conversion rate, and the weighted summation and standardization are general methods for multivariate data analysis;

[0060] S3, based on the drug absorption conversion analysis result and the evolution of the corresponding hepatocellular carcinoma, the future drug conversion component is predicted;

[0061] The specific steps are: S31, obtaining the conversion analysis result of the corresponding drug and the tumor abnormality judgment result on the corresponding conversion path; integrating the drug conversion analysis result (such as metabolite concentration, enzyme activity) and the tumor abnormality score (such as image-based characteristics, mutation load) to construct a multi-modal time series data set, according to: drug metabolism kinetics research shows that the tumor microenvironment (such as hypoxic state) will significantly affect the drug conversion efficiency;

[0062] S32, based on the historical conversion analysis result of the corresponding drug and the tumor abnormality judgment result on the conversion path, the tumor abnormality judgment result on the future conversion path is predicted, wherein the prediction of the tumor abnormality judgment result on the future conversion path is performed by means of deep learning neural network;

[0063] The specific steps are: obtaining the historical conversion analysis result of the corresponding drug and the tumor abnormality judgment result on the conversion path, dividing the obtained historical data into 85% weight, bias training set and 15% weight, bias test set; inputting the 85% weight, bias training set into the deep learning neural network model for training to obtain an initial deep learning neural network model; testing the initial deep learning neural network model using the 15% weight, bias test set, and outputting the initial deep learning neural network model output satisfying the maximum tumor abnormality judgment accuracy on the future conversion path as the deep learning neural network model;

[0064] S33, obtaining the ratio of the predicted tumor abnormality judgment result on the future conversion path to the current tumor abnormality judgment result, obtaining the future drug conversion analysis prediction result by dividing the corresponding drug conversion analysis result by the corresponding ratio, and taking the ratio of the future tumor abnormality score to the current tumor abnormality score as a correction factor of the drug conversion efficiency; according to: the tumor load change is negatively correlated with the drug metabolism clearance rate; advantage: adjusting the prediction result by dynamic ratio to avoid misjudgment caused by static threshold;

[0065] S4, based on the future drug conversion component prediction, the drug treatment effect is warned;

[0066] The specific steps are: comparing the future drug conversion analysis prediction result with the set drug conversion analysis threshold value, if the corresponding future drug conversion analysis prediction result is greater than or equal to the set drug conversion analysis threshold value, it means that the drug effect is appropriate, and no drug replacement warning is needed, if the corresponding future drug conversion analysis prediction result is less than the set drug conversion analysis threshold value, it means that the drug effect is not appropriate, and drug replacement warning is needed, comparing the prediction result with the drug conversion threshold value, triggering the warning, the threshold value is usually obtained based on the historical pharmacokinetics or clinical drug resistance threshold value, and the prediction of the decrease of the metabolic efficiency can trigger the drug replacement in advance;

[0067] The value of the set parameters (such as weight and threshold value) of the embodiment is obtained as follows: the evolution of hepatocellular carcinoma of various patients, the composition of the drug and the composition of the metabolite are obtained, the corresponding drug replacement time is obtained, at the same time, the information of the hepatocellular carcinoma patients whose survival time is greater than or equal to the average value is introduced into each step of the application, whether the drug needs to be replaced is judged, and whether the patient really replaces the drug is obtained, the real result and the judgment result are introduced into the matlab fitting software for continuous data fitting, and the value of the set parameters that meet the maximum judgment accuracy is obtained.

[0068] The benefits of the embodiment are: by integrating the evolution data of hepatocellular carcinoma, the drug composition and the metabolite dynamics, combining deep learning to construct a multi-modal time series prediction model, realizing dynamic correlation analysis of drug metabolism pathway and tumor abnormal score, correcting the first-pass effect through mass spectrometry imaging and enzyme activity, quantifying the inhibition effect of tumor microenvironment on drug conversion, and dynamically adjusting the prediction by using the future tumor abnormal score ratio, the accuracy of the drug replacement timing is improved.

[0069] Embodiment 2

[0070] As shown in Figure 4 The embodiment provides a hepatocellular carcinoma treatment effect analysis system based on metabolic comprehensive analysis, which comprises the following modules:

[0071] A data acquisition module is configured to acquire the evolution of hepatocellular carcinoma, the composition of the drug and the composition of the metabolite.

[0072] A drug absorption and conversion analysis module is configured to perform drug absorption and conversion analysis based on the marker of the composition of the drug and the composition of the metabolite.

[0073] A future drug conversion component prediction module is configured to predict future drug conversion components based on the drug absorption and conversion analysis result and the evolution of hepatocellular carcinoma.

[0074] A treatment effect warning module is configured to perform drug treatment effect warning based on the future drug conversion component prediction.

[0075] The steps of implementing the respective functions of the parameters and the unit modules in the liver cancer treatment effect analysis system based on metabolic comprehensive analysis of the present application can refer to the parameters and steps in the embodiments of the liver cancer treatment effect analysis method based on metabolic comprehensive analysis described above, and will not be described here.

[0076] Embodiment 3

[0077] The electronic device of the embodiment of the present application comprises a processor and a memory, wherein the memory stores computer programs that can be called by the processor, and the processor executes the liver cancer treatment effect analysis method based on metabolic comprehensive analysis by calling the computer programs stored in the memory. It should be noted that all computer programs of the liver cancer treatment effect analysis method based on metabolic comprehensive analysis are implemented by using C language, wherein the data acquisition module, the postoperative brain pathology complexity analysis module, the recovery state dynamic analysis module, the wake-up promotion risk assessment module, the wake-up promotion intervention scheme adjustment module, and the control module are all controlled by a remote server.

[0078] In the description of the present specification, the description referring to the terms "one embodiment", "an example", "a specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0079] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details and do not limit the present application to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A system for analyzing the therapeutic effect of hepatocellular carcinoma based on comprehensive metabolic analysis, characterized in that, The system includes: The data acquisition module acquires information on the evolution of hepatocellular carcinoma, the composition of drugs, and the composition of metabolites. The drug absorption and transformation analysis module performs drug absorption and transformation analysis based on the labeling of the drug's composition and the composition of its metabolites. The future drug conversion component prediction module predicts the future drug conversion components based on the drug absorption and conversion analysis results and the corresponding hepatocellular carcinoma evolution. The prediction of future drug conversion components specifically includes the following steps: S31. Obtain the conversion analysis results of the corresponding drug and the tumor abnormality judgment results on the corresponding conversion pathway; S32. Based on the historical conversion analysis results of corresponding drugs and the tumor anomaly judgment results on the conversion pathway, predict the tumor anomaly judgment results on the future conversion pathway. The prediction of the tumor anomaly judgment results on the future conversion pathway is performed by a deep learning neural network. The specific steps are as follows: Obtain the historical conversion analysis results of corresponding drugs and the tumor anomaly judgment results on the conversion pathway; divide the obtained historical data into an 85% weighted and biased training set and a 15% weighted and biased test set; input the 85% weighted and biased training set into a deep learning neural network model for training to obtain an initial deep learning neural network model; use the 15% weighted and biased test set to test the initial deep learning neural network model, and output the initial deep learning neural network model output that satisfies the maximum accuracy of tumor anomaly judgment on the future conversion pathway as the deep learning neural network model; S33. Obtain the ratio of the predicted tumor anomaly judgment result on the future transformation path to the current tumor anomaly judgment result, and obtain the future drug transformation analysis prediction result by dividing the corresponding drug transformation analysis result by the corresponding ratio. The treatment efficacy early warning module provides early warnings of drug treatment efficacy based on predictions of future drug transformation components.

2. The hepatocellular carcinoma treatment efficacy analysis system based on metabolic comprehensive analysis according to claim 1, characterized in that, The drug absorption and conversion analysis includes the following specific steps: S21. Obtain the composition of the drug and the composition of the metabolites, obtain the composition of the consumed drug, subtract the corresponding remaining components in the metabolites from the composition of each component of the consumed drug to obtain the composition of the consumed drug, obtain the distribution of various consumed drug components in different locations of the liver, and obtain the conversion pathway of the consumed drug composition and the generated metabolites based on the constructed enzyme reaction-based metabolic pathway. S22. Simultaneously acquire the tumor evolution status of various liver tissues, and make tumor abnormality judgment based on the size, mutation burden and image changes of the tumor evolution status of each tissue; S23. Analyze the drug conversion and absorption of corresponding pathways by assessing drug consumption and tumor abnormalities in the corresponding tissues.

3. The hepatocellular carcinoma treatment efficacy analysis system based on metabolic comprehensive analysis according to claim 2, characterized in that, The method of predicting drug treatment efficacy based on future drug transformation components includes the following specific steps: The system compares the predicted results of future drug conversion analysis with the set drug conversion analysis threshold. If the predicted result is greater than or equal to the threshold, the drug is considered effective and no drug replacement warning is needed. If the predicted result is less than the threshold, the drug is considered ineffective and a drug replacement warning is needed. The system compares the predicted result with the threshold to trigger the warning. The threshold is based on historical pharmacodynamics or clinical drug resistance thresholds. Predicting a decrease in metabolic efficiency triggers an early drug replacement.

4. The hepatocellular carcinoma treatment efficacy analysis system based on metabolic comprehensive analysis according to claim 3, characterized in that, The evolution of hepatocellular carcinoma in step S1 includes images before and after drug application. The tumor evolution trajectory is tracked through continuous imaging examinations to obtain changes in the tumor and the metabolic characteristics of the tumor microenvironment, including lactate and glutamine levels. The composition of the drug includes the composition, concentration, half-life, and tissue of action of the drug. The composition of the metabolites includes the drug concentration in the metabolites and the concentration of the components formed after drug transformation in the corresponding metabolites.

5. The hepatocellular carcinoma treatment efficacy analysis system based on metabolic comprehensive analysis according to claim 2, characterized in that, The tumor abnormality assessment includes the following specific steps: The size, mutation burden, and image changes of tumors in each tissue are obtained. The image changes are the average difference between the pixel values ​​of each pixel and the pixel values ​​of normal tissue. The average value is divided by the corresponding standard value to obtain the standardized value. The weighted sum is then used to obtain the corresponding tumor abnormality judgment result.

6. The hepatocellular carcinoma treatment efficacy analysis system based on metabolic comprehensive analysis according to claim 2, characterized in that, The drug conversion and absorption analysis of the aforementioned conversion pathway includes the following specific steps: Obtain the tumor abnormality assessment results of each tissue along the drug-to-metabolite conversion pathway, and the corresponding drug consumption; The tumor abnormality judgment results of each tissue are weighted and summed to obtain the tumor abnormality judgment results on the transformation pathway. The weighting is allocated according to the proportion of drug transformation amount in each tissue to the total drug transformation amount. The drug consumption coefficient is obtained by dividing the drug consumption amount by the corresponding standard value of drug consumption. The corresponding drug conversion analysis result is obtained by dividing the drug consumption coefficient by the tumor abnormality judgment result.

Images